Skip to main content

Full text of "Observations on the geology and zoology of Abyssinia, made during the progress of the British expedition to that country in 1867-68"

See other formats 


Google 


This  is  a  digital  copy  of  a  book  that  was  preserved  for  generations  on  library  shelves  before  it  was  carefully  scanned  by  Google  as  part  of  a  project 

to  make  the  world's  books  discoverable  online. 

It  has  survived  long  enough  for  the  copyright  to  expire  and  the  book  to  enter  the  public  domain.  A  public  domain  book  is  one  that  was  never  subject 

to  copyright  or  whose  legal  copyright  term  has  expired.  Whether  a  book  is  in  the  public  domain  may  vary  country  to  country.  Public  domain  books 

are  our  gateways  to  the  past,  representing  a  wealth  of  history,  culture  and  knowledge  that's  often  difficult  to  discover. 

Marks,  notations  and  other  maiginalia  present  in  the  original  volume  will  appear  in  this  file  -  a  reminder  of  this  book's  long  journey  from  the 

publisher  to  a  library  and  finally  to  you. 

Usage  guidelines 

Google  is  proud  to  partner  with  libraries  to  digitize  public  domain  materials  and  make  them  widely  accessible.  Public  domain  books  belong  to  the 
public  and  we  are  merely  their  custodians.  Nevertheless,  this  work  is  expensive,  so  in  order  to  keep  providing  tliis  resource,  we  liave  taken  steps  to 
prevent  abuse  by  commercial  parties,  including  placing  technical  restrictions  on  automated  querying. 
We  also  ask  that  you: 

+  Make  non-commercial  use  of  the  files  We  designed  Google  Book  Search  for  use  by  individuals,  and  we  request  that  you  use  these  files  for 
personal,  non-commercial  purposes. 

+  Refrain  fivm  automated  querying  Do  not  send  automated  queries  of  any  sort  to  Google's  system:  If  you  are  conducting  research  on  machine 
translation,  optical  character  recognition  or  other  areas  where  access  to  a  large  amount  of  text  is  helpful,  please  contact  us.  We  encourage  the 
use  of  public  domain  materials  for  these  purposes  and  may  be  able  to  help. 

+  Maintain  attributionTht  GoogXt  "watermark"  you  see  on  each  file  is  essential  for  in  forming  people  about  this  project  and  helping  them  find 
additional  materials  through  Google  Book  Search.  Please  do  not  remove  it. 

+  Keep  it  legal  Whatever  your  use,  remember  that  you  are  responsible  for  ensuring  that  what  you  are  doing  is  legal.  Do  not  assume  that  just 
because  we  believe  a  book  is  in  the  public  domain  for  users  in  the  United  States,  that  the  work  is  also  in  the  public  domain  for  users  in  other 
countries.  Whether  a  book  is  still  in  copyright  varies  from  country  to  country,  and  we  can't  offer  guidance  on  whether  any  specific  use  of 
any  specific  book  is  allowed.  Please  do  not  assume  that  a  book's  appearance  in  Google  Book  Search  means  it  can  be  used  in  any  manner 
anywhere  in  the  world.  Copyright  infringement  liabili^  can  be  quite  severe. 

About  Google  Book  Search 

Google's  mission  is  to  organize  the  world's  information  and  to  make  it  universally  accessible  and  useful.   Google  Book  Search  helps  readers 
discover  the  world's  books  while  helping  authors  and  publishers  reach  new  audiences.  You  can  search  through  the  full  text  of  this  book  on  the  web 

at|http: //books  .google  .com/I 


556. 


3 


OBSERYATIONS 


02V  THS 


GEOLOGY  AND  ZOOLOGY  OF  ABYSSINIA. 


;! 

>  I 

■  ♦ 


ii 


..^ 


Qg. 


VATIONS 


LOGY  AND  ZOOLOGY 


OF 


ABYSSINIA, 


MADE  DURING  THE  PBOGRESS  OF  THE  BRITISH 
EXPEDITION  TO  THAT  COUNTRY  IN  1867-68. 


BY 

W.  T.  BLANFORD, 

AstociaU  of  the  Roffol  School  of  Jtfm»;  FeJlow  of  Hyt  GvAoqicaX  Society  of  London  : 

C^yrrtsponding  Member  of  the  Zoological  Society  of  London,  awl  of  the  Isie,  Dresden  : 

Deputy  Superintendent  of  tJhe  Geological  Survey  of  India ; 

LATC  OJEOLOOIBT  TO  THE  ABYaSINIAM   EXPEDITION. 


WITH  ILLUSTRATIONS  AND  GEOLOGICAL  MAP. 


•      » 


•    •  <        «  a 


•    • 


»       • 


•       •    .    • 


*       -  *  V  » 


MACMILLAN  AND   CO. 

1870. 


/7^  if/^j*/  <f/'  Translation  and  Rfjyrnduction  is  re9frvtA,\ 


/ 


,.  0 


f  ^ 


^cjb 


OBSEBVATIONS 

G^LOGI  AND  ZOOLOGY 


k 


OF 


ABYSSINIA, 


MADE  DUBING  THE  PB0GBE8S  OF  THE  BRITISH 
EXPEDITION  TO  THAT  COUNTBY  IN  1867-68. 


BY 

W.  T.  BLANFORD, 

AuoeiaU  ofthsRoffol Schcol cf  mS^;  Fellow  ofthi  GtoVoffical  SoeUty  of  London/ 

CoTre$ponding  Membtr  of  the  Zoological  Society  of  London,  and  of  the  Iti$,  Dntden  ; 

Deputy  SuperinUndejU  of  (A«  Geological  Survey  of  IndAa ; 

LATS  OBOLOOIBT  TO  THE  ABTSSIMIAM   EZPKDITION. 


4W 

WITH  ILLUSTRATIONS  AND  GEOLOGICAL  MAP. 


<  • 
«  »       »     ■       •        • 


*  t,        ■  ' 


bonbon : 
MACMILLAN  AND  CO. 

1 870. 

[The  Right  of  Traiulafion  and  XejrriKhieHon  is  rrterved.] 


3()9^H)3 


U>ND«>N  : 

R.   CLAY,  soys,   AND  TAVI/)R,    I'RrNTKKS, 

DRKAD   STRKET  HIT.L. 


•  •  • 


•  _  _ 


PEEPACE 


The  present  work  contains  an  account  of  the  Geolo- 
gical and  Zoological  Observations  made  by  the  author  in 
Abyssinia,  when  accompanying  the  British  army  on  its 
march  to  Magdala  and  back  in  1868,  and  during  a  short 
journey  in  Northern  Abyssinia,  after  the  departure  of 
the  troops. 

The  book  is  divided  into  three  parts.  The  first  of 
these  contains  a  brief  description  of  the  journey,  and  of 
some  of  the  principal  Geological  and  Zoological  features 
of  the  countries  visited.  The  second  part  is  devoted  to 
Geology ;  it  comprises  a  brief  notice  of  the  observations 
made  by  previous  explorers,  a  general  account  of  the 
formations  examined,  and  a  few  remarks  on  the  geolo- 
gical configuration  of  the  country.  The  third  part  con- 
tains the  Zoological  observations,  and  consists  of  an 
enumeration  of  the  various  animals  collected,  with 
remarks  on  their  habits,  distribution,  &c.,  preceded  by  a 


tl  PREFACE, 

short  account  of  the  principal  works  on  Abyssinian 
Zoology  hitherto  published,  and  some  brief  obsen^ations 
on  the  relations  and  distribution  of  the  Abyssinian 
fauna.  A  few  species  of  Vertebrata  which  appear  to 
have  escaped  the  researches  of  previous  explorers  are 
described  and  figured. 

In  the  map  compiled  by  Messrs.  Stanford  and  Co. 
which  accompanies  the  Geological  part,  an  attempt  has 
been  made  to  represent  the  physical  features  of  the 
Abyssinian  highlands  in  a  more  definite  manner  than 
has  previously  been  done,  and  to  distinguish  between 
the  great  eastern  scarp  of  the  plateau  and  the  minor 
irregularities  of  its  surface.^ 

Although  the  time  spent  in  Africa  by  the  author 
was  only  eight  months,  the  advantages  which  he 
enjoyed  in  consequence  of  being  especially  deputed  by 
the  Indian  Government  enabled  him  to  make  a  con- 
siderable series  of  observations  in  both  the  branches  of 
science  to  which  his  attention  was  directed,  and  to 
obtain  a  valuable  collection,  comprising  more  than  1,600 
specin^ns  of  Vertebrata  alone,  representing  about  360 
species,  besides  Mollusca,  and  a  few  Articulata. 

The  time  which  has  elapsed  since  his  return  from 
Abyssinia  to  India  in  September  1868,  has  been  prin- 

1  This  has  been  done  at  the  miggestion  of  Mr.  Trelawny  Saunders,  and  in 
a  great  measurer  by  his  aid. 


PREFACE.  vii 

cipally  devoted  to  the  examination  and  comparison  of 
these  collections.  The  Government  of  India  has  not 
only  most  liberally  relieved  him  from  his  duties  on  the 
Greological  Survey,  in  order  to  enable  him  to  prepare 
the  present  work,  but  has  also  permitted  him  to  com- 
plete the  same  in  England,  in  order  to  afford  him 
access  to  the  museums  and  libraries  of  Europe. 

The  Government  of  Bombay,  under  whose  orders  the 
writer  was  acting  whilst  in  Abyssinia,  and  for  some 
time  subsequently,  has  also  afforded  him  every  facility 
and  aid. 


CONTENTS. 


PAET  I. 

PAQB 

Pkrsonal  Narratitk 1 


PART  II. 

GEOLOGY, 
Introduction 143 

ncnoif 

I. — Physical  Geography  of  Abyssinia,  and  its  Relations  to  the 

Geology,   with    Remarks    on   DenudatioM   and    on   Lake 

Ashangi 151 

II. — List  of  Geological  Fonnations 163 

III. — Metamorphics 164 

rV.—Adigrat  Sandstones 170 

V. — Antalo  Limestones 176 

VI. — Trappean  Series 181 

VIL — Aden  Series  of  Volcanic  Rocks 190 

VIII. — Recent  Formations 194 

Appendix. — Descriptions  of    the  New  Species  of  Fossils   from    the 

Antalo  Limestone 199 

h 


CONTENTS. 

PART     III. 
ZOOLOGY. 

PAOI 

Introduction 207 

Mammalia 222 

AvES 286 

Rbptilia 444 

Pisces 460 

mollusoa 462 


Index 479 


LIST  OF   ILLUSTRATIONS. 


LITHOGRAPHS. 

To  fam  paQ9 

View  of  the  Amba  or  Hill-Fort  op  Daoa,  and  of  the  neioh- 
BouRiNo  Mountains,  taken  from  the  British  Camping- 
Ground  AT  Belago Coloured,    Frontispiea 

Geological  View  of  the  Portion  of  Abyssinia  traversed  by 
the  British  Expedition  in  1868,  from  Anneslet  Bat  to 
Magdala,  and  of  the  Country  between  Massowa  and 
the  Anseba  Valley Coloured,     143 

PLATK 

I. — 1.    Gazella    DORCAS.       1».    DiTTo  ♦.      2.    G.    Bennetti. 

3.   G.  ARABiCA.      4.   G.  subgutturosa.      5.   G.  Spekei. 

5*.  Ditto  ? 261 

II. — HiRUNDO  iETHiopicA  (W.  Blanf.) Coloured,     347 

III. — 1.  Eremomela  griseo-flava  (TIeugl.)    .    .    .    .  \ 

[Coloured,     365 
2.  Phylloscopus  abyssinicus  (W.  Blanf.)   .     .     .  ) 

IV. — Ruticilla(?)  fuscicaudata  (W.  Blanf.)      .     .    .  Coloured,  369 

V. — Pratincola  semitorquata  (Heugl.) Coloured,  366 

VI. — Alauda  PR.ETERMISSA  (W.  Blan£) Coloured,  388 

VII. — Crithagra  flavivertex  (W.  Blanf.)      ....  Coloured.  414 

VIII. — 1.    HeMICIDARIS  ABYSSINICA.       2.  MODIOLA   IMBRICARIA  (Sow. 

var.),  3.  Mytilus  (?Modiola)  tigrensis.  4.  Pholadomya 
granulifera,  5.  Pholadomya  sublirata.  6.  Ceromya 
faucilirata 199 


PART  I. 


PERSONAL  NARRATIVE 


I  .' 


I 


CONTENTS. 

PART     III. 
ZOOLOGY. 

PAOI 

iNTRODUcrriOK 207 

Mammalia 222 

AvES 285 

Rbptilia 444 

Pisces 460 

mollusoa 462 

Index 479 


^ 


PART  I. 


PERSONAL  NARRATIVE 


B 


xu 


LIST  OF  WOOD  ENQRAFINOS, 


WOOD  ENGRAVINGS. 

FULL-PAGE.  To  /act 

PLAn  pay 

I. — View  or  the  British  Camp  at  Senate,  from  tue  North  35 

II. — View  of  the  Hamas  Valley,  west  of  Senafe    ....  41 

III. — View  of  the  Plateaux  and  Valleys  west  of  Fokada  .  57 

IV. — View  of  Lake  Ashanqi,  from  the  North 77 

VIGNETTES. 

TAOR 

Secdon  of  Trach^'te,  Sandstone,  and  Metamorphic  Beds 42 

Section  of  the  Halai  and  Asawat  Plateaux 50 

Section  of  Belations  between  Trap,  Limestone,  Sandstone,  and  Meta- 

morphics,  near  Dongolo 172 

The  Head  and  Horns  of  a  young  Female  Rhinoceros 244 

The  Horns,  with  a  portion  of  the  Skull,  of  an  adult  Female  Rhinoceros, 

with  the  back  view  of  the  hinder  Horn 245 

Sections  of  the  Agama  aunectans 447 

Sections  of  the  La^/?rfa  MZTii^arica 451 

Sections  of  the  Laccrta  Sturti 463 

Sections  of  the  Acanthodactylus  mucronatus 454 


PERSONAL  NARRATIVE. 


I  LEFT  Bombay  harbour  for  Abyssinia  on  the  4th  of 
December,  1867,  in  the  good  ship  Bucentaur,  C-aptain 
Babot,  one  of  the  large  floet  of  transports  engaged  by 
the  Government  for  the  expedition.  At  the  first  notice 
of  the  expedition  being  ordered,  in  the  middle  of  August, 
I  had  applied  to  the  Government  of  India  to  be  allowed 
to  accompany  it  as  geologist,  an  application  which  was 
immediately  granted ;  but,  owing  to  the  large  staff  of 
officers  urgently  required  at  the  seat  of  war,  and  the 
comparatively  small  number  of  vessels  which  were  started 
in  October  and  November,  I  was  not  able  to  obtain  a 
passage  earlier.  The  vessel  in  which  I  sailed  carried 
commissariat  "  followers," — clerkp,  bakers,  butchers, 
dhooly-bearers,  &c.,  from  all  parts  of  India,  Parsees 
from  Bombay,  Mahrattas  from  the  Dekkan,  Teloogoos 
from  the  Northern  Circars,  Tamuls  from  Madras,  and 
Hindustanis  from  the  North-west  Provinces,  under  the 
command  of  Major  Bardin,  the  chief  of  the  excellent 

b2 


rOFAGE  TO  ANNESLSr  BAY^JDEN.  5 

off  Cape  Fartak,  I  shot  a  peregrine  falcon.  The  next 
day  another  falcon,  probably  also  a  peregrine,  alighted 
on  the  rigging,  but  I  failed  to  secure  it,  and  it  flew  away 
to  seaward. 

Our  vessel  lay  at  Aden  for  a  day  and  a  half,  taking  in 
a  supply  of  water.  I  had  one  ramble  amongst  the 
barren  volcanic  hills,  and,  rather  to  my  surprise,  found 
two  land,  shells,  the  widely-spread  Pupa  insularis^  Ehr. 
(P.  pulla  or  Bulimv3  puUtiSy  Gray),  which  ranges  from 
the  Red  Sea  to  India,  and  which  is  even  found  in  the 
Irawady  valley,  near  Ava ;  and  a  variety  of  Bulimics 
lahiosiis,  MiilL  The  shells  found  were  dead,  and  it  is  not 
easy  to  conceive  upon  what  the  living  molluscs  subsist, 
since  the  spot  at  which  I  found  them,  far  up  the  sides  of 
the  mountain,  is  utterly  destitute  of  vegetation. 

All  the  rocks  around  Aden  harbour  are  of  volcanic 
origin,  as  are  indeed,  so  far  as  is  kno^vn,  all  the  hills 
along  both  shores  of  the  southern  part  of  the  Red  Sea. 
These  hills  are  frequently  isolated  by  plains  of  sand ;  to 
the  north  of  Aden  such  a  plain  extends  for  from  thirty 
to  forty  miles,  far  beyond  Lahej.  That  the  Aden  hills 
consist  of  the  remnant  of  one  or  more  volcanic  cones  is 
amply  evident,  and  the  hollow  in  which  the  cantonment 
is  built  is  usually  spoken  of  as  "  the  crater ;"  but  whether 
it  be  really  in  the  position  of  the  ancient  vent  is  by  no 
means  certain,  the  rocks  having  been  so  greatly  removed 
by  denudation  that  it  is  very  difficult  to  form  any  opinion 
as  to  the  original  form  of  the  cone. 

Aden  is  so  well  known  to  every  Anglo-Indian 
that  any  remarks   upon  its  geology  or  fauna  may  be 


PERSONAL  NAURATirE. 


m        • 
•  _  • 


.'.'•  commissariat  officers  cletaclicd  by  the  Madras  Govern- 
ment to  the  aid  of  the  hai-dly-pressed  Bombay  Commis- 
sariat, and  an  old  friend  of  my  own,  our  acquaintance 
having  begun  many  years  before  in  the  remote  regions 
of  Upper  Pegu. 

We  had  a  very  pleasant  and  most  uneventful  sail  of 
ten  days  to  Aden.  The  pace  was  in  general  too  good 
for  a  towing-net  to  be  used ;  but,  after  some  diffi- 
culty, and  the  loss  of  several  nets,  I  succeeded  to 
some  extent,  and  obtained  three  species  of  lanthina 
(/.  fragilisy  var.  roseola^  Rv.,  L  glohosay  and  /. 
bifida),  two  of  Ilyalcea  {11.  luicinata  and  //.  loiigi- 
rostris),  Styliola  cormfo)*mis,  and  some  of  the  peculiar 
embryonic  forms  Sinmigera  or  Cheletropu  and  Mac- 
gillivraya.  Two  species  of  Litiopa  and  an  Atlanta 
were  also  captiu'cd. 

Numerous  peculiar  forms  of  Crustacea  were  taken, 
besides  endless  Velellw  and  PorpitcB.  It  was  curious  to 
note  how  the  two  latter  and  the  floating  lanthincB  were 
distributed,  as  it  were  in  patches,  over  the  surftice  of  the 
ocean,  especially  towards  the  entrance  of  the  Gulf  of  Aden : 
frequently  after  passing  for  hours  almost  without  seeing 
any,  the  vessel  would  traverse  for  some  miles  a  tract 
abounding  with  floating  life. 

The  Indian  Ocean  is  singularly  poor  in  birds. 
Tropic,  or,  as  the  sailors  call  them,  boatswain  birds 
{Phaeton  ruhricauda),  were  the  only  kind  commonly 
seen.  A  nightjar  came  on  board  about  half-way  between 
Bombay  and  Aden,  of  a  species  which  I  could  not 
identify  ;  and  when  about  forty  or  fifty  miles  from  land, 


VOYAGE  TO  ANNESLEY  BAY-^AJUSE,  7 

I  landed  the  next  day  on  Ajus6  with  some  difficulty, 
the  boat  being  nearly  capsized  in  the  breakers.  The 
island  is  quite  flat,  not  more  than  twenty  or  thirty  feet 
above  high-water  mark,  and  composed  of  a  calcareous 
rock  formed  by  the  consolidation  of  blocks  of  coral  and 
shells.  The  edges  are  quite  vertical  and  cliff'-like, 
although  of  course  only  twenty  or  thirty  feet  high ;  a 
shoal,  over  which  there  are  only  two  or  three  feet  of 
water  at  low  tide,  runs  out  for  a  varying  distance,  from 
a  few  yards  to  a  quarter  of  a  mile,  and  terminates 
abruptly.  Outside  the  reef  the  water  is  generally  about 
ten  to  twelve  fathoms  in  depth.  Other  islands  in  the 
neighbourhood  have  precisely  the  same  formation.  On 
the  flat  stony  surface  of  the  island  stunted  bushes  are 
thinly  scattered,  and  afibrd  sustenance  to  some  of  the 
most  miserable-looking  goats  it  was  ever  my  fate  to  set 
eyes  upon,  dwarfed  in  sfciture,  wretchedly  thin,  and, 
apparently  without  exception,  affected  with  chronic 
catarrh. 

On  the  evening  of  December  21st,  two  steamers  came 
in  from  Bombay  and  Aden,  each  with  a  ship  in  tow, 
and  next  morning  one  of  the  steamers  took  our  vessel 
also,  and  all  started  for  our  destination,  which  we 
reached  in  the  evening.  The  course  lay  round  the  Buri 
peninsula,  through  comparatively  narrow  channels  be- 
tween coral  islands  similar  to  Ajus6,  until  turning 
sharply  to  the  south,  we  entered  the  great  lake-like 
expanse  of  Ghubbet  Daknoo,  or  Annesley  Bay,  as  it  was 
named  by  Lord  Valentia,  through  the  narrow  eastern 
channel  between  the  little  gneiss  island  of  Dissee  and 


H  PERSONAL  NARRATIVE. 

thought  superfluous.  Yet,  like  many  places  frequently 
visited,  it  is  very  rarely,  if  ever,  carefully  examined,  and 
its  fauna  especially  might  be  much  more  completely 
ascertained  with  advantage.  Except  Neophrons,  kites, 
and  a  few  crows  {?Corvi(s  umhrinus),  the  only  other 
common  birds  about  the  hill  in  December  were  Sdxicola 
{Cercoviela)  melanura  and  another  Saxicola,  doubtless 
S.  isahellina.  I  also  saw  a  crag  martin  {Cotyle  rupestris). 
We  left  Aden  on  the  evening  of  the  16th  December, 
and  early  the  next  morning  ran  with  a  fair  wind  through 
the  "Gate  of  Weeping"  into  the  Red  Sea.  In  the 
evening  of  the  18th  we  were  off  Amphila,  but  light 
winds  setting  in,  we  only  reached  the  flat  coral  islands  of 
Ajus^,  forty  miles  further,  on  the  20th.  Here  we  found 
two  or  three  other  vessels  waiting  to  be  towed  to  the 
anchorage  at  Annesley  Bay,  while  H.M.S.  Star  was 
lying  in  apparently  most  unpleasant  proximity  to  the 
breakers,  though  really  in  deep  water;  the  officers  and 
crew  being  busily  engaged  in  completing  a  temporary 
lighthouse.  Another  had  already  been  finished  nearer 
to  the  entrance  into  Annesley  Bay.  A  singular  circum- 
stance had  taken  place  a  day  or  two  before  we  arrived. 
A  small  boat  fix>m  the  ship  had  attempted  to  anchor 
close  to  the  island,  when  the  anchor  was  seized  and  the 
boat  dragged  violently  along  by  an  enormous  fish  ;  evi- 
dently, from  the  men's  description,  a  huge  ray.  Whetlicr 
the  anchor  had  caught  in  the  fish,  or  whether  the  latter 
had  swallowed  the  anchor,  could  not  be  told.  To  save 
the  boat  the  men  on  board  were  obliged  to  cut  the  rope 
attached. 


ANNKSLEr  BJr.  9 

mass,  of  wLicli  the  greater  portion  bas  been  removed  by 
denudation. 

An  account  of  one  or  two  short  excursions  made  from 
the  camp  may  serve  to  give  a  slight  idea  of  the  country 
around  the  bay,  and  its  fauna.  On  the  28th  December 
J  joined  a  small  shooting-party,  consisting  of  Captain 
Daniels  of  the  Transport  Train,  Captain  Chrystie  of  the 
Engineers,  and  Lieutenant  Protheroe  of  the  Madras 
Sappers.  We  left  at  nightfall  in  a  small  Arab  boat  for 
the  head  of  the  bay,  about  eight  or  ten  miles  distant. 
It  was  a  calm  night  when  we  started,  but  a  strong  wind 
sprang  up,  and  we  w^ere  obliged  to  anchor  ojBf  the  oppo- 
site shore,  with  an  uncomfortably  rough  sea.  Our  boat's 
cables,  however,  were  none  of  the  best :  two  parted,  one 
after  the  other ;  the  waves  meantime,  to  make  matters 
pleasant,  breaking  on  some  rocks  just  astern  of  us ;  and 
after  drifting  nearly  into  the  breakers,  our  Arab  crew, 
with  that  amount  of  noise  which  only  Orientals  can 
make  when  they  consider  themselves  in  danger,  suc- 
ceeded in  getting  up  sail  and  beating  oflf.  We  tacked 
about  until  daylight.  In  the  morning  we  had  to  wade, 
more  than  waist-deep,  to  the  shore ;  however,  we  suc- 
ceeded in  landing  our  guns,  and  found  ourselves  on  a 
very  rocky  coast  quite  at  the  bottom  of  the  bay.  There 
were  no  cliffs  of  any  height,  but  black  basaltic  masses 
rose  forty  or  fifty  feet  from  the  sea  in  most  places,  here 
and  there  receding  into  little  sandy  bays,  with  patches 
of  mangrove  bu&hes.  In  one  of  these  we  landed, 
and  as  soon  as  we  could  started  inland  in  search  of 
game. 


8  PKHSOSAL  NJHRATirt:. 

the  volcanic  mass  of  the  Buri  peninsula.  The  bay  is 
well  known  as  one  of  the  most  lovely  spots  in  the  Red 
Sea.  Hills  rise  around  it  on  all  sides,  and,  to  the 
westward,  range  towei-s  over  range  in  the  gi-eat  scaip 
of  the  Abyssinian  highlands.  The  coast  is  fringed  by 
rich  green  bushes,  giving  a  deceptive  appcaiance  of 
fertility. 

It  is  unnecessary  to  dwell  upon  our  landing,  or  to 
describe  the  camp  of  Malkatto  or  ZuUa,  with  its  dust 
and  flies,  its  polyglot  inhabitants,  its  vast  energy,  and 
its  busy,  excited  population.  All  its  peculiarities  have 
been  related  by  abler  pens,  and  I  would  rather  refer  to 
the  excellent  letters  of  Dr.  Austin  in  the  TimeSy  and  of 
Mr.  Shepherd  in  the  Times  of  India ,  and  to  the  descrip- 
tions in  Mr.  C.  Markham's  "  Histor}'-  of  the  Abyssinian 
Expedition,"  and  other  published  accounts,  than  attempt 
to  repeat  what  would  ceitainly  not,  at  this  time,  have 
the  merit  of  novelty.  I  remained  in  ZuUa  and  its 
neighbourhood  until  the  1 0th  of  January,  making  a  few 
excursions ;  and  on  this  and  subsequent  occasions  I  had 
a  fair  opportunity  of  studying  the  rocks  and  fiiuna  of 
the  surrounding  country.  The  camp  itself  stood  upon 
an  extensive  sandy  plain  of  alluvial  origin,  which  ex- 
tends for  several  miles  along  the  coast,  and  inland  as 
far  as  the  base  of  the  gi-eat  ranges,  interru])ted,  however, 
here  and  there,  by  small  craggy  hills.  Two  of  these  ai-e 
about  two  miles  south  of  the  spot  occupied  by  the  camp, 
and  consist  of  a  succession  of  lava  flows  and  scoriaceous 
ash  beds,  much  disturl)cd  and  very  irregular  in  dip, 
being  evidently  merely  the  remains  of  a  large  volcanic 


ANNESLEr  BArSUOOTIXG  EXCURSION.  \  1 

ouly  one  was  kiUed.  We  also  saw  numerous  traces  of 
wild  elephants,  but  tliey  were  said  only  to  visit  this  part 
of  the  country  after  the  spring  rains. 

Passing  over  a  little  rocky  range  of  basaltic  hills  and 
traversing  some  stony  ground,  we  entered  another  small 
plain,  on  which  the  bushes  were  less  thick,  and  saw 
a  numerous  herd  of  a  kind  of  large  gazelle  {Gazella 
Soemmeringil)  y  of  which  two  bucks  soon  fell  to  our  rifles. 
They  were  fine  animals,  equal  in  size  to  the  common 
antelope  of  India,  but  with  longer  legs.  Both  males  and 
females  had  rather  short  lyre-shaped  horns. 

It  was  now  getting  hot,  and  we  had  shot  over  some 
miles  of  ground,  so  we  began  to  return  towards  our  boat. 
On  our  way,  however,  which  lay  over  undulating,  rocky 
ground,  we  met  with  new  kinds  of  game.  Several 
Beni  Israel  {Neotragus  saltianus),  one  of  the  very 
smallest  antelopes  known,  a  most  elegant  animal,  of  an 
ash-grey  colour  on  the  back,  passing  into  rufous  on  the 
flanks  and  legs,  and  white  beneath,  and  short  straight 
horns  in  the  male  only,  ran  out  from  the  bushes  like 
hares,  but  at  fai-  greater  speed.  I  saw  a  few  red- wattled 
lapwings  (Sarcioj)Jiorus  tectus),  and  a  large  flock  of 
painted  sand  grouse  (Pterocles  Liclitensteini).  While 
scrambling  over  the  rocks  close  to  the  shore  I  suddenly 
startled  a  number  of  little  animals  about  the  size  of 
small  rabbits,  which  scuttled  over  a  pile  of  stones  and 
hid  in  the  crevices.  I  rolled  one  over,  and  on  picking 
him  up  recognised  a  small  hyrax,  one  of  the  most  sin- 
gular mammals  known,  with  the  body  of  a  rabbit,  the 
head  of  a  marmot,  no  tail,  and  the  dentition  of  a  rhino- 


10  PERSOXJL  NJRRATirE 

We  had  not  left  the  sliore  before  one  of  the  men  with 
us  descried  a  pig,  and  stalking  up  1  caught  sight  of  a 
beast  certainly  more  curious  than  beautiful.  A  pair  of 
immense  tusks  issued  from  a  huge  misshapen  head,  with 
great  knobs  protruding  at  the  side,  and  a  dispropor- 
tionately broad  muzzle.  The  body  was  not  unlike  that 
of  other  wild  pigs,  except  that  the  tail  was  longer,  with 
a  larger  tuft  of  hair  at  the  end.  A  bullet  rolled  him 
into  the  bottom  of  a  ravine  for  the  moment,  but  he 
ultimately  escaped,  his  courage  being  evidently  by  no 
means  equal  to  his  highly  ferocious  appearance.  It  was 
an  -Mian's  wart-hog  {Phacochcerus  ^liani),  a  form  of 
pig  peculiar  to  the  African  continent,  and  bearing  a  most 
comical  resemblance  to  a  hippopotamus  in  its  enormously 
broad  head. 

Shortly  after  a  small  herd  of  five  or  six  smaller  wart- 
hogs  passed  near  us.  They  kept  in  single  file,  with  their 
tails  straight  in  the  air.  We  then  turned  up  a  nearly 
level  plain,  covered  with  bushes  and  thorny  acacia  trees, 
and  soon  came  upon  an  abundance  of  smaller  game.  A 
large  partridge,  with  the  skin  of  the  throat  naked,  and 
of  a  bright  red  or  orange  colour  {Pternestes  ruhricollis)^ 
rose  in  coveys.  Two  or  three  hares,  with  veiy  Large  ears 
and  long  legs,  but  small  bodies  {Lepus  cegyptlus)^  were 
turned  up  imd  shot.  The  next  game  which  appeared 
was  a  large  flock  of  guinea-fowl  with  blue  wattles 
{Numida 2>tilorhync]ii),  about  two  hundred  in  number, 
which  however  ran  ahead  of  us,  keeping  out  of  shot, 
until  we  succeeded  in  surrounding  them  amongst  some 
bushes.     Several  bustards  {Otis  Arabs)  were  seen,  but 


ANNESLEY  BAY— FAUNA,  13 

alha  and  Budytes  melanocephala),  wliite-breasted  crows 
{Corvus  scapulatxis),  kites  {Milvvs  migrans),  the  small 
red-headed  vulture  {Neophron pileatus),  and  the  scavenger 
vulture  (iV.  percnoptents).  A  little  gi-een  bee-eater 
{Merops  viridissimus),  very  closely  resembling  the  bird 
so  abundant  in  India,  was  common  in  the  mangroves  on 
the  shore ;  and  at  the  hot  spring  of  Atzfut  I  obtained  the 
handsome  red  M.  nuhicus.  The  shore  abounded  with 
gulls  (Lcirus  fuscuSy  L.  Hemprichii,  L.  leucophthahnusy 
&c.),  pelicans  (P.  phillipensis),  terns  of  several  species, 
ring  plovers,  curlews,  egrets,  stints,  and  sandpipers,  and 
flocks  of  Dromas  ardeola  were  not  uncommon.  It  will 
not  a  little  surprise  an  ornithologist  to  observe  that 
several  of  the  birds  mentioned  are  comparatively  rare 
species,  much  less  known  than  those  of  the  less  accessible 
Abyssinian  highlands. 

Farther  inland,  amongst  the  thorny  acacia  trees, 
the  fauna  was  more  varied.  A  most  lovely  little 
Nectarinia  {N.  hcdjessimca,  H.  and  E.)  was  fre- 
quently seen,  but  at  this  season  was  the  only  species 
met  with.  The  red-breasted  bush  shrike  {Laniarius 
cruentatus),  the  long-tailed  robin  (Cercotrichas  erythrop- 
teru8)y  the  black-tailed  chat  {Cercomela  melanura),  a 
weaver  bird  {Hyplmntornis  galbtda),  and  two  species 
of  avadavats  {Pytelia  citerior  and  Estrelda  rliodopyga)^ 
were  some  of  the  kinds  most  frequently  seen. 

Mammals  were  not  numerous.  Hyaenas  were  con- 
stantly heard  at  night ;  they  frequently  came  into  the 
camp,  and  occasionally  attacked  the  mules.  I  never  saw 
one  close  to  the  coast  in  daylight,  but  I  believe  that  the 


12  PERSONAL  NARRATIVE, 

ceros,  to  which  utterly  dissimilar  animal,  strange  to  say, 
this  little  "  dweller  amongst  the  rocks "  has  been  found 
to  be  closely  allied.  To  what  particular  species  the 
specimen  (the  only  one  shot  by  me  in  the  immediate 
vicinity  of  Annesley  Bay)  belongs,  it  is  rather  difficult 
to  say ;  but  I  shall  enter  into  the  subject  in  the  part 
especially  devoted  to  Zoology. 

We  re-entered  our  boat,  and  beat  slowly  back  to  the 
camp  against  a  light  sea-breeze.  Soon  after  our  visit 
it  became  generally  known  that  game  abounded  at  the 
head  of  the  bay,  and,  in  consequence,  so  many  shooting- 
parties  visited  the  spot  that  the  animals  became  too 
wild  to  allow  any  one  to  approach  within  a  reasonable 
distance. 

For  several  days  I  remained  at  the  Malkatto  camp, 
and  occupied  myself  in  collecting  specimens  of  birds,  &c. 
The  most  common  land  birds  were  larks,  chats,  and 
shrikes.  Of  larks,  no  less  than  four  species  were  com- 
mon,  Galerita  cristata,  Caleiidrella  hrachyductyla^  Cer- 
ihilauda  desertorum,  and,  perhaps  most  abundant  of  all, 
a  pretty  little  finch-like  Pyrrhulavda  (P.  melanauchen^ 
Cab.).  Two  SaxicolcB  abounded,  S.  deserti  and  S.  isabel- 
Una.  Two  shrikes  also  were  common ;  and,  singularly 
enough,  both  appear  to  belong  to  species  hitherto  almost 
overlooked :  the  one  a  variety  or  local  race  of  the  Indian 
Laniiis  lahtora,  for  which  Dr.  Finsch  proposes  the 
name  of  L.  fallax,  aud  the  other,  the  rare  L.  isdbeU 
Unas  of  Hemprich  and  Ehreuberg.  A  small  DrymcBca 
{D.  gracilis,  Elipp.)  hid  itself  amongst  the  bushes.  The 
oidy  other  common  laud  birds  were  wagtails  {Motacilla 


MAIWII  TO  UADODJ.  15 

not  uucommon,  but,  as  is  the  wont  of  tlie  group  to 
which  it  belongs,  it  was  much  more  solitary  in  its 
habits  than  its  congener.  It  is  almost  identical  with 
the  common  "chinkara"  or  "ravine  deer"  of -India 
{Gazella  Bennetti). 

The  only  common  reptiles  were  a  small  lizard  (a 
species  of  Acanthodactylus),  and  a  very  venomous  little 
viperine  snake  {Echis  arenicoki),  several  of  which  were 
killed  in  the  camp,  and  which  were  at  one  time  so  pre- 
valent that  it  is  surprising  no  accidents  occurred  from 
their  bite.  All  seen  by  me  were,  however,  too  small  to 
have  rendered  it  probable  that  they  could  have  caused 
death  to  a  man.  Their  presence  in  the  camp,  though 
decidedly  objectionable,  was  not  much  noticed  amongst 
the  rather  numerous  annoyances  which  life  at  Zulla 
entailed. 

On  the  1st  January,  1868,  I  rode  over  to  Hadoda, 
sixteen  miles  distant  to  the  westward,  at  the  entrance 
of  the  Haddas  ravine,  up  which  one  of  the  best- 
known  passes  leads  to  Halai,  Dixa,  and  Takonda  or 
Degonta,  on  the  Abyssinian  highlands.  Water  occurs 
here,  and  the  post,  together  with  another,  four  miles 
further  north,  at  Ooah  or  Wia,  had  been  occupied  by 
strong  detachments  of  the  3d  Bombay  Cavalry,  the 
majority  of  whom,  however,  in  consequence  of  great 
mortality  amongst  their  horses,  had  already  been  re- 
moved to  the  highlands,  leaving  only  a  small  guard  of 
dismounted  sowars  in  charge  of  the  commissariat  stores, 
tents,  &c.  Major  Bardin  having  been  sent  to  arrange 
about  the  removal  of  surplus  stores,  I  accompanied  him. 


14  PERSONAL  NARIijrirK. 

spotted  species  (//.  crocnta)  Wiis  that  most  prevalent. 
//.  stinata  is  also  said  to  occur.  Jackals  were  common 
enough,  but  by  some  singular  fiitality  I  could  never 
succeed  in  securing  a  specimen.  When  on  horsel>ack 
without  a  gun,  1  have  often  passed  within  a  few  yai-ds  ; 
but  when  looking  for  specimens,  I  rarely  saw  more  than 
a  tail  retreating  at  full  speed  amongst  the  bushes  fifty 
yards  away.  All  I  can  state  of  the  kind  is  tliat  it  is 
quite  different  from  the  jackal  of  India,  and  also  from 
that  of  the  Abyssinian  highlands,  being  a  nmch  more 
slightly-built  animal,  witli  longer  legs  and  a  mucli 
greyer  colour.  The  cry,  too,  is  very  distinct  It  is 
probably  Canis  riparius  of  Hemprich  and  Ehrenberg; 
but  species  of  jackals,  like  cats,  IL/races,  and  many 
other  animals,  have  been  confused  by  naturalists,  many 
of  whom  only  know  them  in  the  form  of  dried  skins, 
to  an  extent  which  renders  identification  difficult. 
I  once  obtained  a  fine  wild  cat  {F.  maniculata), 
which  took  refuge  in  the  camp  amongst  a  pile  of  pack- 
saddles,  and  was  caught,  singularly  enough,  on  one  of 
the  hooks  intended  for  holding  the  loads.  Hares  were 
common  until  they  met  their  fate  at  the  hands  of  the 
British  soldier  and  his  dog,  and  the  only  mammals  except 
the  hyaenas  and  jackals  which  profited  by  tlie  British 
occupation  were  some  small  bright  iTifous  jerboa  mice, 
with  very  long  hind  legs,  which  found  unwonted  supplies 
of  food  in  the  commissariat  stores,  and  increased  and 
multiplied  until  the  ground  around  the  huts  and  tents 
was  riddled  with  their  holes.  Besides  the  large  Soemmer- 
ing's  gazelle  previously  mentioned,  the  Dorcas  gazelle  was 


HALODA^BABOONS.  1 7 

wonder  that  the  men  were  not  sanguine.  They  had  lost 
their  horses  by  the  epidemic  which  at  that  time  raged 
amongst  the  arrimals,  and  they  were  left  isolated,  whilst 
their  fellow-soldiers  had  proceeded  to  the  front  No 
human  being  has  fewer  resources  in  himself  than  the 
native  of  India.  Many  of  them  were  suffering  from 
fever,  mainly  caused,  I  expect,  by  disgust  They  were 
removed  and  sent  to  the  front  shortly  after. 

On  rising  the  next  morning  I  saw  a  singular  spectacle. 
A  large  troop  of  baboons,  at  least  200  in  number,  were 
hunting  for  any  corn  dropped  upon  the  ground  in  the  place 
where  the  horses  had  been  picketed.  They  were  the  first 
I  had  seen,  though  the  sight  of  these  uncouth  monkeys 
soon  became  familiar  enough.  The  species  (Cynocephalics 
hamadryas)  is  the  well-known  dog-faced  baboon  of 
North-eastern  Africa  and  Arabia,  the  same  which  is 
frequently  represented  on  Egyptian  monuments.  The 
male  is  a  most  formidable-looking  animal,  something 
between  a  lion  and  a  French  poodle  in  appearance,  with 
long  hair  over  his  shoulders  and  foreparts.  The  day 
before  they  had  come  into  the  commissariat  enclosure, 
and  commenced  pilfering  the  grain. 

The  road  from  Hadoda  to  Wia  was  uninteresting.  It 
traversed  an  undulating  plain  of  metamorphic  rocks, 
with  thin  jungle  of  thorny  acacia  trees.  On  the  next  day 
I  returned  to  ZuUa,  but  before  doing  so  I  climbed  a  high 
hill  close  to  Hadoda,  in  order  to  make  out,  so  far  as 
possible,  the  general  strike  of  the  rocks,  which,  as  is 
commonly  the  case  in  metamorphic  countries,  especially 
where  the  dip  is  low,  appeared  irregular.     On  my  way 


16  PERS(h\JL  NARRJTirE. 

The  road  was  pretty  well  marked  across  the  plain, 
through  a  narrow  gorge  between  basalt  rocks,  where  the 
Haddas  has  cut  its  way  through  the  little  spur  of 
volcanic  beds  stretching  south  from  the  Gadam  range, 
and  thence  along  the  dry  sandy  watercourse,  to  the  base 
of  the  hills.  Along  the  banks  of  this  watercourse  are 
some -fine  tamarisk  trees,  aflFording  shelter  to  numerous 

•  

bulbuls  {Pycnoiiotus  arsinoe),  hombills  {Tockus  ery- 
throrliynchus),  barbets  {Trachyphonus  margaritatus)^ 
and  other  birds,  whilst  under  their  shade  the  red-throated 
francolins  {Pternestes  rtihincollis),  Beni  Israel  antelopes, 
and  hares  abounded,  and  on  the  surrounding  plain  we 
saw  a  few  of  both  kinds  of  gazelle.  I  shot  a  doe  Soem- 
mering*s  gazelle,  and  could  have  killed  another,  but; 
having  no  means  of  carrying  the  animal  in,  I  let  it  go. 
At  the  base  of  the  main  mass  of  hills  the  road  entered 
upon  metamorphic  rocks,  and  traversed  a  small  gorge 
cut  through  them,  in  which,  for  the  first  time  since  land- 
ing on  the  African  coast,  our  eyes  were  gladdened  with 
the  sight  of  running  water,  and  immc^diately  after  we 
reached  the  camp  at  Hadoda. 

It  was  far  from  a  lively  sight.  A  few  very  dusty 
tents  were  surrounded  by  a  thoni  fence,  and  some  very 
miserable-looking  Indian  native  troopers  lounged  about, 
backed  by  a  huge  herd  of  camels,  the  most  helpless, 
stupid,  and  melancholy-looking  of  all  the  beasts  of  the 
universe.  A  dismounted  dragoon  is  proverbial,  and 
certainly  a  dismounted  sowar,  if  those  who  garrisoned 
the  posts  at  Hadoda  and  Wia  be  taken  as  fair  specimens, 
is  not  an  improvement  on  the  European  prototype.     No 


HALODA^BABOONS.  1 7 

wonder  that  the  men  were  not  sanguine.  They  had  lost 
their  horses  by  the  epidemic  which  at  that  time  raged 
amongst  the  airimals,  and  they  were  left  isolated,  whilst 
their  fellow-soldiers  had  proceeded  to  the  front  No 
human  being  has  fewer  resources  in  himself  than  the 
native  of  India.  Many  of  them  were  suffering  from 
fever,  mainly  caused,  I  expect,  by  disgust.  They  were 
removed  and  sent  to  the  front  shortly  after. 

On  rismg  the  next  morning  I  saw  a  singular  spectacle. 
A  large  troop  of  baboons,  at  least  200  in  number,  were 
hunting  for  any  corn  dropped  upon  the  ground  in  the  place 
where  the  horses  had  been  picketed.  They  were  the  first 
I  had  seen,  though  the  sight  of  these  imcouth  monkeys 
soon  became  familiar  enough.  The  species  {Cynocephalus 
hctmadryas)  is  the  well-known  dog-faced  baboon  of 
North-eastern  Africa  and  Arabia,  the  same  which  is 
frequently  represented  on  Egyptian  monuments.  The 
male  is  a  most  formidable-looking  animal,  something 
between  a  lion  and  a  French  poodle  in  appearance,  with 
long  hair  over  his  shoulders  and  foreparts.  The  day 
before  they  had  come  into  the  commissariat  enclosure, 
and  commenced  pilfering  the  grain. 

The  road  from  Hadoda  to  Wia  was  uninteresting.  It 
traversed  an  undulating  plain  of  metamorphic  rocks, 
with  thin  jungle  of  thorny  acacia  trees.  On  the  next  day 
I  returned  to  Zulla,  but  before  doing  so  I  climbed  a  high 
hill  close  to  Hadoda,  in  order  to  make  out,  so  far  as 
possible,  the  general  strike  of  the  rocks,  which,  as  is 
commonly  the  case  in  metamorphic  countries,  especially 
where  the  dip  is  low,  appeared  irregular.     On  my  way 

c 


1 8  PERSONAL  NARRATIVE, 

up  the  hill  a  small  animal,  apparently  a  squirrel  which 
had  lost  part  of  its  tail  by  accident,  ran  into  a  hole 
amonsrst  some  rocks  before  I  could  shoot  it.  As  I  was 
sitting  down  on  the  top  of  the  hill  taking  some  bearings, 
I  saw  another,  and  presently  a  third,  all  with  equally 
short  tails,  on  some  very  precipitous  rocks,  and  creeping 
down  cautiously  I  succeeded  in  shooting  both ;  but  after 
having,  with  no  small  difficulty,  climbed  down  a  rather 
ugly  piece  of  cliff  to  the  spot,  I  could  only  find  one.  It 
was  evidently  no  squirrel  :  its  rather  differently  shaped 
head,  singularly  soft  close  fur,  and  short  tail,  which  was 
evidently  natural  and  not  accidental,  showed  it  to  be 
some  other  animal ;  and  on  comparing  it  afterwards 
with  Mr.  Blyth's  description,  I  was  delighted  to  find 
that  I  had  rediscovered  the  rare  rodent  Pectinator 
Spekeiy  first  found  by  the  late  gallant  and  distinguished 
African  traveller,  after  whom  it  is  named,  in  his  explora- 
tion of  the  Somali  country  in  1854. 

The  remainder  of  my  stay  in  Zulla  or  Malkatto  camp 
was  principally  occupied  in  collecting  animals  and  ex- 
ploring the  geology.  Although  there  was  no  risk  whatever 
in  riding  about  the  country,  there  was  a  great  drawback 
in  the  difficulty  of  procuring  water,  and  the  necessity 
for  horses  to  be  in  camp  morning  and  evening,  as 
otherwise  they  ran  the  risk  of  going  without.  The 
commissariat  arrangements  also,  which  necessitated 
the  drawing  up  of  a  long  document,  as  elaborate 
and  formal  as  a  mercantile  invoice,  before  any  rations 
could  be  obtained,  involved  a  considerable  loss  of 
time  at  first. 


QEOLOOICAL  FEATURES  OF  ANNESLET  BAY,  \  9 

Before  proceeding  to  describe  any  part  of  my  subse- 
quent journey,  a  brief  sketch  of  some  of  the  geological 
features  observed  in  the  neighbourhood  of  the  landing- 
place  may  be  given. 

The  eastern  shore  of  Annesley  Bay  appears  to  consist 
mainly^  if  not  entirely,  of  volcanic  rocks.  The  island  of 
Dissi,  however,  at  the  entrance  of  the  bay,  is  entirely 
composed  of  metamorphics,  the  strike  of  which  is  nearly 
north  and  south. 

The  high  range  of  Gadam,  the  peaks  of  which  are 
about  3,000  feet  above  the  sea,  rises  sharply  from  the 
west  shore  of  the  bay  near  its  entrance,  and  is  entirely 
composed  of  metamorphic  rocks,  with  the  same  general 
strike  as  Dissi  The  hill  is  comparatively  isolated,  the 
usual  road  from  Zulla  to  Arkiko  passing  behind  it  At 
the  base  of  the  main  range  of  the  Abyssinian  highlands 
a  few  miles  farther  west,  the  rocks  near  Hadoda,  at  the 
entrance  to  the  Haddas  pass,  consist  of  mica-schist  and 
homblend  schist.  At  Wia  granitoid  gneiss  occurs. 
These  rocks  have  a  low  dip  to  west-south-west,  variable 
in  amount,  but  not  exceeding  30°  or  40^ 

Between  Hadoda  and  the  southern  portion  of  the 
Gadam  range  there  is  an  alluvial  plain  covered  with  sand 
and  gravel,  evidently  deposited  by  the  torrents  from  the 
hills.  Between  this  and  the  coast  plain  around  Zulla 
a  low  spur  extends  for  some  miles  from  the  south  end 
of  the  Gadam  range,  composed  of  volcanic  rocks,  similar 
to  those  forming  the  low  hills  already  mentioned 
near  Zulla.  Through  these  the  Haddas  torrent  cuts 
a  small    gorge,  with    precipitous  basaltic    cliflfs    of  no 

c3 


20  PERSONAL  NARRATIVE. 

great  height  on  each  side.  The  road  from  Malkatto  to 
Hadoda  passed,  as  already  mentioned,  through  this 
gorge  ;  the  beds  intersected  appear  to  be  horizontal ; 
farther  south,  at  the  extreme  end  of  the  spur,  they  dip 
east  at  about  20^  but  the  dip  is  not  constant,  and 
there  has  been  much  disturbance.  One  stratum,  a  little 
north  of  the  road  from  Zulla  to  Komayli,  consists  of 
soft  lapilloB  ;  it  is  stratified,  and  presents  the  appearance 
of  having  been  deposited  from  water,  as  it  contains  flakes 
of  mica  and  fragments,  rounded  apparently,  of  volcanic 
ash.  None  of  these  hills  present  the  appearance  of 
having  been  parts  of  a  distinct  cone. 

The  plain  formed  by  the  alluvial  deposits  of  the 
Had  das,  the  Komayli  stream,  and  another  torrent  further 
south,  extends  along  the  coast  for  about  eight  miles 
south  of  Zulla,  as  far  as  the  hot  spring  of  Atfd  or 
Atzfut,  which  rises  at  the  edge  of  a  dense  cluster  of  man- 
grove trees,  close  to  the  seashore.  The  water  is  brackish, 
but  not  sufficiently  so  to  prevent  its  being  drunk  by 
animals.  The  rocks  which  here  approach  the  shore  are 
volcanic,  and  similar  .beds  skirt  the  hills  for  some  dis- 
tance to  the  west  Farther  inland,  gneissic  rocks  crop 
out,  and  the  range  along  the  edge  of  the  bay  to  the 
southward,  as  far  -as  the  Turkish  outpost  at  Arafild,  is 
composed  of  metamorphics.  On  one  hill,  south  of  Atzfut, 
trap  rests  upon  the  gneiss  at  some  height  above  the  base 
of  the  hill.  At  Arafild  trap  comes  in  again,  a  small 
alluvial  plain  intervening  near  the  sea,  on  which,  at  the 
time  of  my  visit,  there  was  an  immense  herd,  400  or 
600  in  number,  of  the  fine  antelope  Gazella  Scemmer' 


KOMAYLI— WATER-SUPPLY.  21 

ingiiy  which  at  first  abounded  around  Zulla  also,  but 
was  gradually  extirpated  or  driven  away.  Just  south  of 
the  Turkish  military  camp  at  Arafild  is  a  very  perfect 
volcanic  cone,  evidently  of  much  later  date  than  the  two 
other  volcanic  hills.  It  is  about  300  feet  high,  and  com- 
posed mainly  of  fine  volcanic  ash. 

On  the  7th  of  January,  at  the  desire  of  Colonel  Wil- 
kins,  I  rode  to  Komayli,  the  first  march  on  the  road  laid 
out  to  the  highlands,  to  examine  the  water-supply.  The 
road  was  similar  to  that  to  Hadoda,  but  much  more 
dusty,  in  consequence  of  the  traffic.  Long  strings  of 
mules  and  camels,  laden  with  stores  of  all  kinds,  were 
constantly  to  be  seen  going  and  returning.  The  railway 
was  as  yet  in  embryo,  the  first  steam-engine  being  landed 
about  this  time.  Komayli  lies  at  the  entrance  of  a  better- 
defined  gorge  than  that  which  debouches  at  Hadoda; 
and  when  the  pioneer  force  first  examined  the  locality, 
there  was  running  water  here  also.  This  had  disap- 
peared ;  but  wells  of  no  great  depth  had  been  dug  in  the 
torrent  bed,  and  the  little  American  Norton's  pumps, 
afterwards  so  largely  used,  had  been  employed  for  the 
first  time.  A  brief  examination  of  the  wells  convinced 
me  that  there  was  very  little  fear  of  the  water  failing ; 
and  I  was  enabled  to  report  most  favourably  on  the 
prospects  of  what  was,  at  the  moment,  one  of  the  most 
important  necessaries  for  the  progress  of  the  army.  My 
report  only  confirmed  that  of  the  engineer  officers,  who 
had  already  examined  the  spot,  and  was  fully  borne  out 
by  the  results.  The  wells  at  Komayli  not  only  supplied 
amply  the   immense  camp  of  both  men  and  animals 


22  PERSONAL  NARRATIVE, 

which  was  concentrated  there,  but  had  the  campaign 
lasted  longer,  it  was  intended  to  have  taken  the  water  in 
pipes  to  Zulla. 

The  temperature  of  the  water  was  remarkably  high, 
so  much  so  that  I  suspected  part  of  the  supply  was 
derived  from  a  hot  spring.  But  all  wells  in  this  neigh- 
bourhood are  singularly  warm.  In  one  which  was  sub- 
sequently dug  by  the  Pimjab  Pioneers,  about  six  miles 
fix)m  Zulla,  the  water  was  far  hotter  than  at  Komayli, 
and  certainly  much  above  the  average  temperature  of  the 
air.  Coupling  these  facts  with  the  presence  of  a  hot 
spring  on  the  shore  at  Atzfut  abeady  mentioned,  and 
with  that  at  Ailat  to  be  subsequently  noticed,  it  appears 
probable  that  nearly  all  the  subterranean  waters  in  the 
neighbourhood  of  Zulla  and  Massowa  have  an  unusually 
high  temperature;  a  fact  perhaps  connected  with  such 
evidence  of  geologically  recent  volcanic  action  as  is 
afforded  by  the  cone  at  Arafil^. 

I  rode  back  to  Zulla  the  same  evening.  The  Com- 
mander-in-chief, Sir  Kobert  (now  Lord)  Napier,  had 
landed  in  the  morning,  and,  on  hearing  my  report, 
directed  me  to  proceed  to  Mayen  or  Undul  Wells,  two 
marches  beyond  Komayli,  and  ascertain  all  I  could  con- 
cerning the  water-supply  there.  The  pass  which  had 
been  selected  for  the  road  to  the  Abyssinian  highlands 
consisted  of  a  ravine  running  nearly  north  and  south, 
or  parallel  to  the  general  direction  of  the  great  scarp 
bordering  the  high  country,  from  Komayli  to  Senaf^,  a 
distance  along  the  bed  of  the  torrent  of  about  fifty  milea 
In  two  places,  at  Sum,  twelve  miles  beyond  Komayli, 


KOMATLL  23 

and  at  Rahaguddy,  or,  as  it  was  more  commonly  called, 
Rereguddy,  ten  miles  below  Senafd,  nmning  water  was 
found ;  but  between  these  two  a  march  of  twenty-nine 
miles  intervened.  In  order  to  break  this,  a  well  had 
been  dug  by  the  engineers  at  a  place  called  by  the 
natives  Mayen,  but  subsequently  known  in  the  army  as 
Undul  Wells  (the  Undul  being  a  torrent  which  enters  the 
main  ravine  a  little  above).  My  particular  duty  was  to 
ascertain,  as  far  as  possible,  the  prospects  of  the  water- 
supply  at  Mayen,  and  to  inquire  if  it  could  be  supple- 
mented in  case  of  failure. 

After  some  difficulty  in  obtaining  mules,  I  started  on 
the  10th,  and  rode  across  to  Komayli.  The  next  morn- 
ing I  started  up  the  pass,  with  Captain  C.  B.  Smith,  of 
the  Commissariat,  and  Lieut.  St.  John  of  the  Engineers, 
who  had  charge  of  the  telegraph  in  course  of  construction. 
The  latter  is  an  excellent  ornithologist,  and  both  on  this 
occasion  and  often  subsequently  gave  me  information 
and  specimens. 

Komayli,  as  already  mentioned,  lies  at  the  spot  where 
the  long  ravine  ends,  on  the  verge  of  tlie  coast  plain. 
The  road,  on  leaving  the  camp,  enters  a  valley,  about 
half  a  mile  broad  at  first,  but  much  narrower  afterwards, 
its  bottom  composed  of  a  deposit  of  boulders,  pebbles, 
and  sand  brought  down  by  the  torrent,  upon  which  a 
tolerably  dense  vegetation  has  sprung  up.  The  hills  rise 
abruptly  from  the  side ;  they  are  entirely  composed  of 
metamorphic  rocks.  Close  to  Komayli,  gneiss  and  mica- 
ceous schist,  with  garnets,  are  seen,  the  lamination 
dipping  east  on  the  south  side  of  the  ravine,  and  north- 


24  PERSONAL  NARRATIVE. 

east  on  the  north,  at  a  low  angle.  About  three  miles 
up  the  ravine  its  course,  which  has  so  far  been  nearly 
due  west,  turns  suddenly  to  the  south,  and  assumes  the 
direction  which,  with  many  local  changes,  continues 
to  Senaf^ ;  and  here  the  foliation  of  the  gneiss  rolls 
over,  and  dips  first  north-north-west  and  then  west 

The  road  at  this  time  was  only  very  imperfectly  made. 
The  process  of  road-making,  by  the  various  regiments 
employed,  consisted  in  marking  out  a  trace  twenty  feet 
broad,  felling  any  trees  which  might  be  in  the  way, 
cutting  down  steep  banks,  and  removing  the  boulders. 
The  road  thus  formed  was,  of  course,  rather  heavy,  being 
often  over  gravel,  but  still  a  great  improvement  on  the 
mere  torrent  bed.  But  in  January  a  commencement 
only  had  been  made,  the  whole  efforts  of  the  Sappers 
being  devoted  at  first  to  rendering  practicable  such  diflB- 
cult  parts  as  the  Sum  defile.  It  was  not  always  very 
easy  to  tell  the  road,  as  lateral  valleys  of  considerable 
size  sometimes  entered  the  main  one.  The  path,  how- 
ever, was  pretty  well  marked  in  general,  and  could 
usually  be  recognised  by  the  carcases  and  skeletons  of 
mules  and  camels  scattered  along  it  We  passed  several 
of  the  Shoho  inhabitants  of  the  country — uncouth-looking 
beings,  mostly  with  bushy  masses  of  hair  transfixed  by 
wooden  or  horn  skewers,  little  or  no  beard,  and  a  small 
quantity  of  very  dirty  cotton  clothing.  The  women  had 
invariably  a  leathern  petticoat,  frequently  ornamented 
with  cowrie  shells.  They  were,  if  an}i;hing,  more  ill- 
favoured  than  their  lords  and  masters.  At  a  subse- 
quent period  of  the   expedition,   these   people  became 


SHOnOS— FAUNA  OF  THE  PASS.  25 

80  far  civilized  as  to  dress  themselves  in  an  outer 
garment  made  of  old  gmmy  bags,  the  coarse  Indian 
fabric  in  which  grain  is  packed.  In  January,  such 
proofs  of  advanced  civilization  had  not  extended  beyond 
ZuUa. 

These  people  are  genuine  Arabs,  their  hands  against 
every  man's,  arrant  thieves  and  cut-throats,  but  still 
with  some  principle  of  honour.  Owing  to  the  influence 
Mr.  Mimzinger  had  with  them,  he  succeeded  in  inducing 
them — for  payment,  of  course — ^not  only  to  cease  from 
plundering  to  a  great  extent,  but  to  become  carriers 
to  the  army  ;  and  enormous  quantities  of  commissariat 
stores  were  conveyed  by  them.  Some  petty  pilfering,  of 
course,  took  place,  but,  I  am  convinced,  far  less  than 
was  generally  supposed. 

There  is  no  cultivation  in  or  about  the  pass,  so  far  as  I 
saw.  The  people  live  mainly  on  the  milk  of  their  herds, 
and  on  "  durra "  or  millet  {jawari)  brought  from  the 
highlands.  The  cattle  are  small,  but  well  formed  and  in 
good  condition,  all  being  humped,  and  having  horns  of 
moderate  size  spreading  from  the  side  of  the  head,  not 
rising  in  a  curve  as  in  Indian  humped  cattle.  The  goats 
are  rather  fine,  the  males  being  magnificent  patriarchal 
fellows,  with  grand  twisted  horns  two  feet  and  upwards 
in  length. 

The  faima  in  the  lower  portion  of  the  pass  was  not 
remarkable.  Vultures,  of  course,  abounded,  and  I  once 
or  twice  saw  a  lammergeyer  {Gypaetos  meridionalis), 
conspicuous  by  bis  pointed  wings  and  tail  and  great  size. 
The  most  interesting   novelty  to   me   was  a  kind  of 


26  PERSON AL  NARRATIFR 

ground  squirrel  {Xeriis  rutilics,  Riipp.),  of  which  we 
saw  several  together  running  over  the  rocks.  It  has 
rather  long,  coarse,  harsh,  almost  bristly  hair,  and  a  fine 
bushy  tail. 

About  six  or  seven  miles  from  Komayli  we  met  a 
native  of  India  with  a  camel  and  a  mule,  who  com- 
plained of  having  been  robbed  by  two  Shohos.  The 
thieves  had  run  into  a  small  ravine  close  by,  and  we 
immediately  afterwards  saw  them  climbing  up  the  rocks. 
I  slipped  a  couple  of  bullet-cartridges  into  my  breech- 
loader, but  my  companions  begged  me  not  to  shoot  the 
thieves,  and  I  felt  no  particular  inclination  to  draw  the 
first  blood  in  the  campaign,  so  I  contented  myself  with 
frightening  the  Shohos  by  sending  five  or  six  bullets  as 
near  to  their  heads  as  I  could  without  risk  of  hitting 
them.  I  was  rather  glad  I  had  let  them  oflF  when 
I  found  that  the  whole  plunder  consisted  of  a  small 
bag  of  rice,  which,  of  course,  they  had  to  abandon  in 
their  flight. 

About  ten  miles  from  Komayli  the  ravine  suddenly 
contracted,  owing  to  the  greater  hardness  of  the  rocks, 
and  we  entered  the  magnificent  gorge  known  as  the  Sum 
pass.  Here  the  formation  consists  of  a  very  hard, 
massive  rock,  chiefly  composed  of  felspar  and  quartz; 
almost  all  trace  of  foliation  is  lost  in  general,  but  where 
it  can  be  made  out,  it  is  generally  nearly  vertical  The 
water,  which  in  other  parts  of  the  ravine  runs  beneath 
the  gravel,  is  brought  to  the  surface,  and  forms  a  little 
stream  containing  small  fish.  The  bed  of  the  torrent 
consists  of  immense  blocks  of  rock,  some  of  which  were 


KOMAYLI  PASS^SUR  U.  27 

being  removed  by  blasting,  whikt  over  others  a  ramp 
was  made  to  carry  the  road.  A  large  number  of  Bombay 
Sappers  and  Beloochees  were  occupied  in  this  work,  and 
the  narrow  gorge  resounded  with  the  hum  of  voices  and 
the  ring  of  hammers  and  crowbars,  as  it  had  certainly 
never  done  before  since  the  torrent  first  cut  it  out  of  the 
solid  rock. 

On  each  side  of  the  ravine  rose  steep  beetling  clifis  to 
a  great  height.  They  are  in  most  places  quite  inaccessible, 
and  on  one  occasion,  in  May,  towards  the  close  of  the 
expedition,  a  flood  came  down  from  the  highlands,  and 
some  mules  and  men  were  swept  away. 

We  reached  the  top  of  this  gorge,  and  foimd  the  camp 
at  Suru,  with  the  Beloochees  under  Major  Bevill.  It  was 
in  a  narrow  cramped  space,  and  uncomfortably  hot  in 
the  middle  of  the  day.  The  lovely  little  Nectarinia 
habessinica  abounded  in  some  small  trees,  and  numerous 
Hyraces  inhabited  the  rocks  around.  I  here  also  again 
came  upon  Pectinator  Spekei^  the  peculiar  rodent  first 
seen  at  Hadoda. 

The  next  morning  Dr.  Boustead,  of  the  Beloochees,  who 
had  been  out  to  look  after  elephants,  returned,  having 
killed  two  in  a  valley  about  ten  miles  away  to  the  east- 
ward. I  was  busy  skinning  two  or  three  squirrels  and 
other  animals  which  I  had  shot  the  day  before,  and  did 
not  start  tiU  mid-day.  Owing  to  the  reports  of  Shoho 
thieving,  I  did  not  like  to  leave  my  mules  alone,  and 
accompanied  them.  Soon  their  wretched  condition  began 
to  telL  One  fell,  and  I  transferred  the  greater  part  of 
the  load  to  my  horse  ;  then  another  dropped ;  and  finally. 


28  PERSONAL  NARRATIVE, 

after  endless  delay,  I  only  reached  the  camp  at  Mayen  or 
Undul  Wells  aft;er  nightfall. 

A  great  change  took  place  in  the  fauna  on  this  part  of 
the  road.  Sum  is  2,000  feet  alvove  the  sea,  Undul  Wells 
about  3,400,  and  the  sub-tropical  fauna  is  entered,  con- 
taining some  of  the  animals  peculiar  to  the  Abyssinian 
highlands.  Instead  of  the  white-breasted  crow  of  Zulla, 
a  black,  short-tailed  bird  {Co'nms  affinis)  occurred.  It  is 
very  raven-like  ;  indeed,  it  is  classed  as  a  raven  by  some 
naturalists.  A  king  crow  (Dicrurus  divaricattts)^  a  very 
noisy  yellow-billed  hombiU  {Tockus  jlavirostris)^  a  crate- 
ropus  (C  leiLCopygius),  a  large  partridge  (Francolinus 
JErkelii),  and  a  very  handsome  bee-eater  (Merops 
Lcifre&nayi)  were  conspicuous. 

The  next  morning  was  devoted  to  the  wells.  Two 
had  been  dug  in  the  gravel  of  the  torrent  bed,  and 
Lieut.  Le  Mesurier,  an  old  acquaintance,  who  was  in 
charge,  was  busily  engaged  enlarging  and  deepening 
them,  and  adapting  one  for  a  kind  of  chain-pump. 
The  rocks  at  Mayen,  though  not  quite  so  hard  ad 
at  Suru,  are  very  massive  and  quartzose,  striking 
north-north-west  and  dipping  at  a  high  angle  to  the 
westward.  They  consist  of  various  kinds  of  gneiss 
and  homblend  schist,  with  quartz  veins  and  a  few 
greenstone  dykes. 

Our  examination  of  the  well  was  interrupted  by  the 
return  of  a  corporal  of  Sappers,  who  had  started  for 
Senaf^,  but  after  going  about  two  miles,  had  been  knocked 
oflF  his  mule  by  some  Shohos,  and  both  his  mule  and  his 
rifle  had  been  carried  off".     A  chase  was  soon  organized, 


GUINEA-FOWL  PLAIN.  29 

but  was  fruitless ;  neither  mule  nor  rifle  was  ever 
seen  again. 

About  a  mile  and  a  half  up  the  ravine  above  Mayen,  at 
a  spot  where  several  smaller  valleys  enter  it,  is  a  little 
plain,  about  a  quarter  of  a  mile  across,  and  rather  more 
than  half  a  mile  long,  tenanted  by  many  guinea-fowl 
when  the  pioneers  of  the  army  passed  by,  and  hence 
always  known  as  Guinea-fowl  Plain.  It  is  a  flat,  fonned 
of  boulders  and  granite  deposited  from  the  streams,  and 
covered  partly  with  thick  thorny  bush-jungle,  partly  with 
a  very  prickly  aloe-like  plant.  It  was  much  haunted  by 
wart-hogs,  hyaenas,  and  Beni  Israel,  until,  like  the  guinea- 
fowls,  all,  except  the  hyaenas,  found  their  way  into  the 
camp  cooking-pots,  and  served  to  eke  out  the  tough  beef 
of  commissariat  rations.  Small  as  it  is,  this  plain  is  the 
broadest  piece  of  level  ground  met  with  in  the  paas,  and 
from  it  a  view  is  obtaiaed  up  some  of  the  side  valleys  ; 
one  of  these,  the  Undul  torrent  already  mentioned,  ex- 
poses at  its  head  some  noble  cliffs  of  saadstone,  part  of 
the  scarp  of  a  plateau  near  Takonda  between  the  Haddas 
and  Komayli  valleys.  Tliis  small  plain  is  also  note- 
worthy as  being  the  first  place  on  the  road  to  the  high- 
lands, where  that  fine  and  remarkable  euphorbiaceous 
plant,  the  kolqual,  was  met  with.  To  Anglo-Indians  this 
plant  was  perhaps  less  striking  than  to  Europeans,  as  a 
closely  allied  species  abounds  in  parts  of  the  Bombay  Presi- 
dency ;  but  the  Indian  plant  is  inferior  in  size  and  beauty. 

It  struck  me  that  the  base  of  these  sandstone  cliffs 
might  be  a  very  likely  spot  for  a  copious  supply  of 
water,  and  that  it  was  at  least  worth  the  trouble  to 


30  PERSONAL  NARRATIFE, 

examine  them.  Inquiry  from  the  Shohos  was  not  satis- 
factory, especially  when  conducted,  as  in  the  present 
instance,  through  the  medium  of  an  Arab  interpreter, 
who  spoke  very  little  Hindustani,  and  an  old  Shoho 
woman,  who  understood  still  less  Arabic.  However, 
having  procured  a  guide,  Le  Mesurier  and  I  started  to 
explore  the  Undul  valley,  taking  with  us  a  small  tent 
and  a  day's  provisions  on  mules. 

We  had  a  most  delightful  and  profitable  ride,  although 
the  quantity  of  water  found  was  small.  For  about  six 
or  eight  miles  we  found  a  well-marked  cattle-track 
leading  up  the  torrent  bed,  at  the  sides  of  which  were 
small  flats,  covered  with  fine  acacia  trees.  Gradually 
the  valley  contracted,  and  the  hiUs  at  the  sides  became 
higher,  and  in  many  places  covered  with  huge  tufts  of 
coarse  grass.  At  first  these  were  all  at  a  considerable 
elevation  above  our  path,  but  we  soon  ascended  to  their 
level.  The  quantity  was  large,  and  the  discovery — ^for  it 
was  a  discovery — most  important,  since  all  forage  had 
hitherto  been  brought  from  the  seaboard,  employing 
many  mules,  and,  in  fact,  taking  up  a  very  large  pro- 
portion of  the  carriage  required  for  transporting  pro- 
visions and  stores  to  the  highlands.^  The  grass  was 
coarse,  but  the  mules,  and  even  the  horses,  ate  it ;  and 
soon  afterwards  arrangements  were  made  by  Sir  Robert 

^  Although  the  mules  were,  at  this  time,  very  poorly  fed,  it  was  calculated 
that  in  going  from  the  camp  at  ZuUa  to  Senaf^  and  back  they  consumed  more 
than  half  their  load.  I  pointed  out  the  grass  in  the  Undul  valley  to  Lieut 
Sturt,  of  the  Transport  Train,  who  brought  it  to  the  notice  of  the  Commander- 
in-chief^  and  was  placed  in  charge  of  the  arrangements  for  collecting  it  by 
means  of  Shohos. 


UNDUL  RAVINE.  31 

Napier,  by  which  all  the  camps  in  the  pass  were  supplied 
from  the  locality  discovered  by  us,  and  others  in  the 
neighbourhood,  thus  eflFecting  an  enormous  gain  in  the 
carrying  power  of  the  transport  animals. 

A  large  number  of  birds  which  I  had  not  previously 
seen  made  their  appearance  a  little  above  5,000  feet. 
Amongst  these  the  most  conspicuous  was  the  Abyssinian 
plantain-eater  {Turacus  leucotis)y  which  flew  from  tree 
to  tree,  high  above  our  heads  for  the  most  part,  dis- 
playing its  gorgeous  crimson  quill-feathers  in  its  flight. 
I  was  somewhat  surprised  at  seeing  a  green  pigeon 
{Tver on  abyssinica)  in  small  flocks.  The  handsome 
francolin  {K  Erkelii )  was  very  abundant ;  at  almost 
every  turn  of  the  valley  one  or  more"  were  seen,  which 
on  our  approach  ran  into  the  bushes,  or  up  the  sides 
of  the  hills,  rarely  taking  to  flight.  This  francolin  is 
a  noble  game  bird,  the  male  being  at  least  equal  to 
a  pheasant  in  size.  A  few  Beni  Israel  showed  them- 
selves, and  several  Hyraces  scampered  away  amongst  the 
rocks;  but  four-footed  animals  were  not  numerous.  A 
considerable  change  took  place  in  the  vegetation,  two 
or  three  kinds  of  Ficus  appearing,  two  of  which  were 
scarcely  distinguishable  from  the  peepul  and  banyan  of 
India.  The  temperature  also,  towards  evening,  was 
decidedly  cool. 

Our  Shoho  guide  had  for  some  time  persisted  that 
there  was  no  possible  road,  and  that  we  should  find  no 
water;  but,  as  there  was  an  excellent  cattle-track,  we 
preferred  seeing  for  ourselves,  and  at  last  he  admitted 
that  there  was  a  very  little  water  in  one  place.     About 


32  PERSONAL  NARRATIVE. 

evening  we  emerged  from  the  narrow  valley  we  had  been 
traversing,  at  the  very  base  of  the  sandstone  cliffs  which 
had  attracted  om*  attention  in  the  distance,  and  found  a 
small  spring  of  excellent  water.  Of  course  this  was  of 
no  practical  value  to  the  aimy  ascending  the  pass:  it 
was  amply  sufficient  for  our  wants,  however,  and  that  of 
our  animals,  as  well  as  supplying  a  Shoho  village  close  by, 
the  people  from  which  brought  us  milk,  and  were  very 
civil  for  Shohos.  They  called  their  village  Undul — a 
name  which  has  been  applied  to  the  camping-ground  in 
the  main  pass,  ten  miles  away. 

The  next  day  we  attempted  to  climb  the  sandstone 
plateau.  It  was  a  most  fatiguing  ascent,  through  bushes 
and  creepers,  most  of  them  thorny,  and  over  very  bad 
ground.  We  reached  the  sandstone,  however ;  but,  finding 
that  another  long  descent  and  ascent  intervened,  did  not 
go  on  to  the  plateau.  I  afterwards  found  that  this  spot 
was  near  Azawat,  nearly  north  of  Takonda.  The  view 
was  very  extensive.  Over  the  ranges  which  hem  in  the 
Senaf^  pjiss  a  portion  of  Armesley  Bay  and  its  eastern 
shore  could  be  seen.  The  extreme  horizon  was  misty; 
but,  looking  to  the  southward,  it  was  just  possible  to 
distinguish  a  vast  sandy  expanse,  with  a  peculiar  whitish 
colour,  the  northern  portion  of  the  great  salt  plain  near 
Amphila. 

The  metamorphic  rocks  extend  for  a  long  distance  up 
the  hiU,  and  are  capped  by  about  500  feet  of  coarsfe 
white  sandstone.  They  are  much  contorted,  but  preserve 
a  general  north  and  south  strike.  Close  to  the  spring  in 
the  valley  are  some  highly  bituminous  schists. 


VALLEYS  AROUND  MAYEK  33 

In  tliis  ramble  I  obtained  a  few  land-shells,  the  first 
I  had  met  with.  There  were  not  many  kinds.  Helix 
Dai^naudiy  Vitrina  liuppelliana,  a  single  specimen  of 
Bulimus  Olivieriy  and  two  small  PupcB,  were  all  I  could 
find.  They  appeared  more  abundant  on  the  upper  parts 
of  the  hill-side. 

We  returned  to  our  camping-groimd,  and  found  it  was 
too  late  to  go  back  to  Undul  Wells  that  day,  so  deferred 
doing  so  till  next  morning.  On  reaching  it,  I  found  one 
of  my  horses  ill  with  the  diflease  which  had  proved  so 
fatal  to  the  cavalry  and  baggage  animals.  For  two  or 
three  days  he  was  most  dangerously  ill,  but  at  length 
recovered,  and  I  subsequently  used  him  throughout  the 
campaign.  Kum-and-water  and  quinine  were  the  re- 
medies used.  At  first  the  horse  objected  to  them ;  but 
after  a  day  or  two  he  took  a  strong  liking  to  grog,  and 
swallowed  his  dose  with  the  greatest  satisfaction. 

I  spent  three  or  four  days  in  searching  the  side  valleys 
around  May  en,  and  ascertaining  the  presence  or  absence 
of  springs.  Small  quantities  of  water  occur  in  several 
instances,  usually  at  a  place  where  a  band  of  very  hard 
compact  rock  crosses  the  valley,  and  causes  a  very  small 
stream,  w^hich  elsewhere  runs  beneath  the  gravel  at  the 
bottom  of  the  ravine,  to  appear  at  the  surface.  There 
is  almost  invariably  a  sudden  drop  in  the  valley  at  this 
spot,  so  that  after  rain  there  must  be  a  waterfall,  and 
in  the  dry  season  the  water  trickles  slowly  over  the 
rocks.  All  the  springs  appear  to  contain  carbonate  of 
lime,  which  is  deposited  in  immense  masses  of  calcareous 
tufa. 


34  PERSONAL  NJRRATIFB, 

One  spot  struck  me  as  very  remarkable.  I  followed  a 
ravine  which  enters  the  main  pass  at  Guinea-fowl  Plain 
for  about  three  miles.  Some  Shohos  tried  to  turn  me 
back,  but,  although  I  was  alone,  I  was  well  armed,  and 
found  it  convenient  to  misunderstand  them.  At  length 
I  reached  a  spot  where  the  valley  appeared  absolutely  to 
terminate.  On  one  side  was  a  precipice  of  granitoid 
gneiss  about  a  hundred  feet  high,  on  the  other  a  still 
liigher  wall  of  tufa  overhanging  at  the  top.  Down  the 
granitic  rock  trickled  the  smallest  of  streams,  sufficient, 
however,  to  attract  numerous  hamadryas  monkeys,  who 
sat  on  the  rocks  around.  The  tufa  cliffs  were  the  roost- 
ing-place  of  a  large  colony  of  Amydrtis  Blythi,  a  rare 
bird,  something  like  a  starling  with  a  long  tail,  black, 
with  chestnut  wings.  In  the  evening  these  birds  were 
flying  around  in  considerable  numbers,  keeping  up  a 
constant  chattering  cry.  I  subsequently  revisited  the 
spot,  and  shot  several  specimens,  to  the  consternation 
of  the  monkeys,  who  howled  at  me  from  the  rocks 
above. 

In  Lieutenant  Sturt,  who  arrived  some  days  before  at 
Undul  Wells  in  charge  of  the  transport  train  at  that 
camp,  I  found  a  capital  ornithologist,  and  we  had  many 
rambles  together  in  search  of  birds.  At  a  subsequent 
period  the  same  officer  took  charge  of  one  of  my  native 
collectors.  The  fauna  of  Undul  Wells  and  its  neigh- 
bourhood is  very  interesting,  comprising  several  birds 
"which  are  found  neither  on  the  highlands  nor  on  the  sea- 
coast,  such  as  two  woodpeckers  {Picus  Hemprichii  and 
P.  nubicns)y  a  little  barbet  {Barhatula  pusilla),  and  a 


i 


I 

i 


I ' 


PJSS  BETWEEN  UXDUL  WELLS  AND  SENAFE,  35 

wren  {Oligocercvs  micrurus)^  almost  all  of  which  I  sub- 
sequently found  in  the  Anseba  valley. 

On  the  21st  of  January  I  rode  up  the  pass  to  Senaf6. 
Except  that  the  ravine  gradually  contracts,  there  is 
no  great  change  in  its  appearance  for  some  miles  above 
Undul  Wella  The  hill-sides  become  greener,  and 
similar  changes  take  place  in  the  fauna  and  flora  to 
those  already  noticed  in  the  Undul  ravine,  plantain- 
eaterSy  green  pigeons,  and  other  birds  making  their 
appearance.  About  Rahaguddy,  at  an  elevation  of 
6,000  feeti  the  alteration  is  much  greater.  The  mcta- 
morphic  rocks  have  become  much  more  slaty,  a  cha- 
racter which  they  maintain  till  beyond  Senaf^.  The 
road  close  to  the  camping-ground  at  Rahaguddy  traverses 
a  very  narrow  gorge  with  a  small  stream  of  water  trick- 
ling through  it,  which  stream,  however,  is  not,  like  that 
at  Suru,  constant  throughout  the  year. 

Ascending  beyond  Eahagiiddy,  the  path  traverses  more 
open  ground,  the  immediate  sides  of  the  ravine  being 
lower  and  less  steep,  so  that  the  hills  beyond  are  visible. 
To  the  south  tower  the  sandstone  precipices  of  Mount 
Sowera;  on  all  other  sides  are  fine  hills,  of  varying 
form,  composed  of  metamorphic  rocks,  sandstone  or  trap, 
covered  with  forest,  still  thin,  but  much  finer  than  that 
in  the  lower  portion  of  the  pass.  Leaving  at  length  the 
torrent  bed,  the  road  ascends  through  what  was,  in 
January,  a  wood  of  high  juniper  trees.  Here  I  found  a 
large  party  of  the  Punjab  Pioneers  engaged  in  making 
the  cart  road,  which  was  finished  soon  after.  It  was 
afternoon,  and  the  air  was  delightfully  rool,  even  cold, 

D   '2 


36  PERSONAL  NARRATIVE. 

producing  the  exhilarating  effect  peculiar  to  a  mountain 
atmosphere,  and  which  is  nowhere  felt  so  thoroughly  as 
within  the  tropics. 

At  the  commencement  of  the  ascent  from  the  ravine, 
trachytic  rock,  approaching  claystone  in  mineral  cha- 
racter, appears,  and  continues  nearly  to  the  top  of  the 
scarp,  where  basalt  replaces  it.  The  trachyte  is  so  much 
decomposed  that  it  resembles  argillaceous  sandstone,  and 
it  was  not  till  afterwards  that  I  distinguished  its  real 
character.  The  opposite  side  of  the  ravine  consists  of 
metamorphic  rocks,  with,  in  one  spot,  an  isolated  block 
of  sandstone. 

I  soon  reached  the  top  of  the  scarp,  and  found  myself 
fairly  on  the  highlands  of  Abyssinia.  A  broad  flat  open 
valley  with  some  fine  isolated  hills  on  each  side  led  to 
Senaf^,  and  I  cantered  pleasantly  along  a  good  even  road. 
The  ground  around  had  been  recently  in  cultivation. 
No  villages  were  to  be  seen  at  first,  but  soon  under  the 
precipitous  sides  of  the  hills  a  number  of  flat  terrace-like 
projections  became  visible,  and  these  on  a  nearer  approach 
proved  to  be  flat-topped  huts,  the  walls  built  of  stones, 
and  so  closely  resembling  the  hill-sides  in  colour  that 
they  might  easily  be  taken  for  a  portion  of  them.  About 
two  miles  from  the  top  of  the  pass  the  valley  expanded 
into  a  plain  dotted  over  with  the  white  tents  of  the 
British  camp.  To  the  right  hand,  w^est  of  the  camp, 
were  a  number  of  barren  craggy  hills  of  most  singular 
form,  apparently  of  sandstone,  but  really  of  claystone 
and  trachyte.  These  hills  will  be  described  in  a  subse- 
quent page. 


RETURN  TO  ZULU.  37 

I  only  remained  a  night  in  Senaf6.  It  was  bitterly 
cold,  and  in  the  morning  the  ground  was  white  with 
hoar-frost.  The  next  day  I  redescended  the  pass,  examin- 
ing on  my  way  several  places,  where  small  quantities  of 
water  were  found.  The  supply  was  in  no  case  sufficient 
to  prove  of  much  ser\'ice  in  supplementing  the  resources 

of  Undul  Wells. 

< 

At  Mayen  I  found  much  improvement.  The  well  had 
been  deepened,  and  an  improved  kind  of  chain  pump 
(Brasyer's)  substituted  for  the  little  American  pumps 
(Norton's)  which  had  hitherto  been  used.  A  careful 
test  of  the  supply  by  Lieut.  Protheroe  and  myself 
showed  that  the  well  could  yield  about  700  gallons 
an  hour. 

After  a  day  or  two  spent  at  Undul  Wells,  I  returned 
once  more  to  the  sea-coast.  I  had  not  been  able  to  take 
any  of  my  appliances  for  collecting  skins,  &c.  with  me, 
and  had  consequently  left  my  native  taxidermist  with  my 
tents  at  Zulla.  On  my  way  down  I  met  Sir  Robert 
Napier,  who  was  on  his  road  to  the  highlands,  and 
I  reported  on  the  prospects  of  water-supply  in  the 
passes. 

I  found  the  camp  at  Zulla  enormously  increased  in 
size.  There  were  now  about  10,000  troops  encamped, 
and  at  least  as  many  camp-followers  of  all  kinds.  A 
splendid  fleet  of  upwards  of  100  large  ships  and 
steamers  lay  at  anchor  in  the  roadstead.  A  large  bazaar 
had  been  established,  and  shops  filled  with  supplies  of 
all  kinds  abounded.  I  found  also  that  Dr.  Cook,  the 
meteorologist,  had  arrived  from  Bombay,  bringing  with 


38  PERSONAL  NJRRATIVK 

him  a  second  native  taxidermist  sent  to  me  by  Dr. 
Anderson,  of  the  Indian  Museum.  The  two  men  had 
collected  a  considerable  number  of  birds,  many  of  them 
of  great  interest. 

With  Dr.  Cook  I  made  another  excursion  of  three  days 
to  the  head  of  the  bay,  and  obtained  several  moUusca  and 
a  few  birds.  The  game  had  become  very  wild.  For 
several  days  I  occupied  myself  geologizing  and  collecting 
around  Zulla.  A  change  took  place  in  the  weather :  a 
little  rain  fell,  and  the  hills  were  covered  with  dense 
masses  of  clouds.  We  heard  that  heavy  rain  was  falling 
almost  daily  about  the  lower  portion  of  the  pass. 
Meantime  troops  and  stores  were  being  rapidly  pushed 
on  to  the  front,  and  the  camp  was  once  more  diminishing 
in  size. 

It  was  not  till  the  12th  February  that  I  could  obtain 
carriage  with  which  to  start  again  for  the  highlands,  and 
even  then  it  was  with  great  difficulty  that  I  could 
manage  to  take  with  me  sufficient  stores  for  collecting. 
I  remained  two  days  at  Komayli,  where  a  considerable 
change  had  taken  place,  grass  having  sprung  up  and  the 
trees  put  forth  fresh  leaves.  The  only  conspicuous 
change  in  the  fauna,  however,  was  the  presence  in  large 
numbers  of  the  black  Abyssinian  crow  {Corvus  affinis), 
which,  a  month  before,  I  had  not  met  with  below  Undul 
Wells.  Subsequently,  several  kinds  of  birds  immigrated, 
many  of  which  were  still  to  be  found  in  May. 

Rain  fell  so  heavily  on  the  road  from  Komayli  to  Suru 
that  my  camels  could  not  get  on,  and  remained  out  all 
night.     This  necessitated  another  day's  halt.     At  Undul 


RE- ASCEND  TO  THE  HIGHLANDS.  39 

I  again  stopped  for  a  couple  of  days  collecting,  and  I 
finally  reached  Senafe  once  more  on  the  21st  February. 
The  rains  did  not  extend  into  the  hills  quite  so  far  as 
Undul  Wells,  and  above  the  weather  was  fine,  but  less 
cool  than  in  January. 

By  this  time  the  advanced  guard  had  reached  Antalo. 
The  Commander-in-chief  was  about  two  marches  beyond 
Adigrat.  I  took  up  my  abode  in  the  commissariat  lines 
near  Major  Thacker's  tent,  one  of  the  best  parts  of  the 
camp,  and  whilst  awaiting  permission  to  go  on  to  the 
front  I  set  to  work  to  examine  the  geology  of  Senafe  and 
the  neighbourhood,  which  was  very  interesting,  and  to 
collect  as  usual. 

I  have  already  briefly  mentioned  the  position  of  ScDaf^ 
in  a  plain  with  hills  at  both  sides.  Beyond  the  craggy 
rocks,  just  west  of  the  camp,  lies  the  head  of  a  deep 
valley.  To  the  north-east  is  the  great  mass  of  Sowera, 
entirely  formed  of  sandstone ;  but  south  of  it,  separated 
by  a  small  valley,  is  a  trachyte  hill.  The  majority  of  the 
other  rises  are  volcanic  in  origin,  consisting  of  basalt  or 
trachyte,  the  latter  greatly  predominating.  The  greatest 
extent  of  these  rocks  is  north-west  of  Senafe,  in  which 
direction  they  stretch  more  than  half-way  to  Takonda. 
The  culminating  point,  Akiib  Teriki  (or  Arabi  Teliki),  is 
a  most  conspicuous  object  from  all  the  country  around. 
It  lies  nearly  due  north  of  Senafe,  and  west  of  the  road 
at  the  head  of  the  Komayli  pass.  From  the  spot  where 
trap  first  appears,  at  the  base  of  the  last  ascent  in  the 
paas,  to  the  top  of  this  hiU,  is  a  height  of  2,000  feet,  and 
the  whole  consists  of  trachyte  or  ba.salt.     It  is  by  no 


40  PERSONAL  NARRATIVE, 

means  clear  whether  this  mass  is  intrusive,  or  whether  it 
fills  a  hollow  in  the  metamorphics. 

The  hill  (Akiib  Teriki)  is  flat  at  the  top,  the  uppermost 
portion  consisting  apparently  of  a  trachyte  lava  flow, 
rather  coarse  and  gritty,  resembling  sandstone.  It  is  of 
a  greyish  colour  on  a  freshly  fractured  surface,  but 
weathers  brown.  It  contains  grains  of  magnetic  iron, 
which  affect  the  compass.  The  lower  portion  of  the  hill 
consists  partly  of  basalt  and  partly  of  trachyte. 

A  band  of  hiUs  to  the  north-west  is  composed  of  hard, 
greenish-black  basalt,  compax^t  and  fine-grained,  contain- 
ing a  Uttle  quartz  and  agate  in  irregular  nodulea 
Beyond  these  again  to  the  west,  and  also  to  the  south, 
there  is  a  considerable  hilly  tract  of  trachyte.  There  is 
nothing  like  bedding  nor  any  regular  superposition  of  one 
rock  upon  the  other,  such  as  is  seen  near  Adigrat  and 
south  of  Antalo  ;  the  basalt  is  seen  in  one  place  about  five 
miles  north-west  of  Senaf^  abutting  against  the  trachyte 
as  if  faulted,  in  another  spot  it  appears  to  emerge  from 
beneath  it.  Some  volcanic  ash  is  seen,  which  adds  to  the 
probability  of  a  volcanic  vent  having  existed  in  the  im- 
mediate neighbourhood,  but  the  rocks  are  far  too  confused 
for  their  history  to  be  made  out  satisfactorily. 

The  trachyte  forms  cliffs  of  great  height  on  the  side  of 
the  deep  valley  west  of  Senafe  and  south-west  of  Ak&h 
Teriki,  the  bottom  of  which  consists  of  metamorphic 
rocks.  (See  Plate  II.)  The  same  rock  forms  a  low  ridge 
bounding  the  little  cultivated  plain  at  Senafe  on  the 
west  It  is  very  compact  and  hard,  greatly  resembling 
a  hardened  sandstone. 


J 


I 


6E0L0QY  OF  COUNTRY  AROUND  SENAFE,  41 

This  ridge  joins  the  Senafe  rocks,  the  remarkable  crags 
already  referred  to  (see  Plate  L),  which  are  immediately 
west  of  the  position  occupied  by  the  British  camp  during 
tlie  campaign.  The  rock  composing  these  hills  is  softer 
and  more  earthy  than  that  of  the  other  rises,  and  the 
resemblance  to  an  argillaceous  sandstone  is  most  remark- 
able. The  form  of  these  hills,  and  of  those  near  Adowa, 
which  are  visible  in  the  distance,  and  may  probably  be 
of  the  same  rock,  is  so  similar  to  the  immense  hummocks 
which  generally  occur  in  granite  and  granitoid  gneiss, 
that  they  might  easily  be  mistaken  at  a  distance  for  the 
latter  formation,  as  they  were  at  the  first  sight  by  Colonel 
Pha3rre  and  the  pioneers  of  the  army.  On  a  nearer 
approach,  their  resemblance  to  sandstone  is  greatly  in- 
creased by  their  brown  colour.  Another  point  in  which 
they  simulate  sedimentary  rocks  is  in  being  traversed  by 
dark  ferruginous  streaks  and  bands. 

The  hill  immediately  east  of  Senafe  is  of  a  rather 
different  trachyte,  greyer  and  more  crystalline.  It  rests 
upon  a  thin  bed  of  sandstone,  and  the  latter  upon 
metamorphics,  so  that  the  volcanic  rock  is  in  this  case 
evidently  a  flow.  As  will  be  seen  presently,  the  same 
is  the  case  further  south.  This  hill  affords  an  excellent 
example  of  the  complete  imconformity  between  the 
trachyte,  even  when  distinctly  occurring  as  a  flow,  and 
the  sandstone  ;  for  although  the  latter  rock  upon  the  sides 
of  this  hill  nowhere  exceeds  fifty  feet  in  thickness,  the 
next  hill  to  the  north,  one  of  the  spurs  of  Sowera,  only 
separated  by  a  valley  not  half  a  mile  broad,  is  entirely 
composed  of  sandstone,  and  is  higher  than  the  trachyte 


PERSOSAL  yjSRJTIFK 


hills,  being  300  or  400  fuet  at  least  above  the  plain,  some- 
what as  represented  in  the  accompanying  sketch  section. 


1,  Trachyte;  2,  Sanditone  ;  3,  Metamorphica. 
Another  hill  similarly  composedj  trachyte  resting  on 
a  thin  bed  of  sandstone,  with  metamorphics  below,  lies  ■ 
about  three  miles  south  of  Senaf^,  just  west  of  the  road 
to  Guna  Guna.  To  the  west  of  this,  and  south  of  the 
deep  ravine  already  mentioned  aa  commencing  just  west 
of  Senaf^,  is  another  extensive  plateau,  mainly  composed 
of  sandstone,  from  which,  about  eight  miles  west  by 
south  from  Seuaf^,  there  rises  a  fine  trachyte  hill,  called 
Kishyat  The  northern  face  of  this  hill  is  precipitous, 
and  consists  of  magnificent  vertical  columns,  which 
extend  the  whole  height  of  the  clifl^,  upwards  of  300 
feet,  without  a  single  curve  or  break.  The  rock  is 
pale  greyish,  or  purplish,  more  crystalline  than  the 
trachytes  near  Senaf^.  Another  small  hiU  of  the  same 
composition  is  met  with  on  the  road  from  Senafd  to 
Kishyat  A  third,  a  little  inferior  only  to  Kishyat 
in  size,  lies  across  the  deep  ravine  to  the  south  of  that 
hiU,  at  a  distance  of  about  five  miles.  Others  are  scat- 
tered over  the  surface  of  the  sandstone  to  the  southward 
near  Fokada,  and  will  be  mentioned  presently.  It  is 
by  no  means  improbable  that  these  are  all  remains  of 
the  same  great  lava  flow. 


SOWERA— FAUNA  OF  SENAFE,  43 

The  sandstone  below  the  trachyte  on  the  hill  east  of 
Senaf(^,  and  that  on  the  other  hill  to  the  south,  may 
possibly  be  distinct  from  the  massive  Adigrat  sandstone, 
but  there  is  no  diflference  in  mineral  character  of  any 
importance.  Beneath  Senafd  rocks  no  sandstone  can  be 
traced,  just  as  beneath  Akiib  Teriki.  Kishyat  appears 
to  rest  on  sandstone. 

The  great  mass  of  Sowera  requires  no  special  descrip- 
tion. It  consists  of  rather  coarse  sandstone,  white  in 
colour,  with  brown  ferruginous  bands  here  and  there. 
Other  similar  great  plateaux  to  the  westward  correspond 
with  it — that  of  Tsaro,  between  the  Komayli  ravine  and 
that  of  the  Haddas,  and  the  southern  portion  of  the 
Halai  plateau  west  of  the  Haddas.  All  of  these  consist 
of  massive  horizontal  beds  of  sandstone,  and  are  bounded 
by  huge  precipices.  They  are,  in  fact,  great  flat-topped 
spurs,  projecting  from  the  mass  of  the  table-land,  and 
separating  the  different  valleys. 

The  fauna  of  the  neighbourhood  of  Senaf^  is  quite 
different  from  that  in  the  pass  below,  and  is  fairly  tj^ical 
of  that  inhabiting  the  temperate  regions  of  Abyssinia. 
Some  of  the  mammals,  however,  as  the  hamadryas 
monkey,  the  spotted  hyaena,  and  the  wart-hog,^  are 
identical  with  those  found  on  the  sea-coast ;  others,  as 
the  hyrax  {H.  Brucei),  the  ichneumon  {Herpestes  mut- 
gigella),  the  klipspringer,  and  koodoo,  extend  into  the 
sub-tropical  region,  whilst  a  few  appear  peculiar  to  the 
highlands  proper.     Amongst  these,  the  principal  are  the 

*  As  I  did  not  shoot  this  myself,  nor  see  it  close  enough  to  be  quite  sure 
of  its  identity,  I  am  not  certain  of  the  animal  being  the  same. 


44  PERSONAL  NARRATIVE, 

jackal  (Ccuiis  variegatus,  Riipp.),  the  hare,  apparently 
an  undeseribed  species,^  and  the  striped  ground-squirrel 
{Xerus  leuco-umhriniis).  The  latter  1  subsequently  met 
with  in  the  Anseba  valley,  but  it  does  not  appear  in 
the  passes  below  Senaf^  at  a  lower  level  than  about 
6,000  feet. 

I  obtained  a  good  series  of  the  birds.  Of  the  Raptores, 
falcons  were  scarce,  with  the  exception  of  the  kestrel ; 
of  eagles,  Aquila  rapax^  closely  allied,  both  in  appear- 
ance and  habits,  to  the  "  wokab  "  of  India,  was  abundant, 
and  the  noble  **  Bateleur  "  {Hehtarsus  ecaudatus)  was 
occasionally  seen  soaring  at  a  great  height,  its  long 
pointed  wings,  pure  white  beneath,  contrasting  with  its 
black  body.  The  beautiful  white-breasted  buzzard  {Buteo 
augur)  was  far  from  rare.  Around  the  Ciunp  kites 
{Milvus  migrans  and  M.  cegyptius)  and  vultures  {Gyps 
Rdppelli)  abounded,  and  the  Abyssinian  lammergeyer 
{GypaHtos  meridionalis)  was  perpetually  hovering  or 
sweeping  with  his  powerful  flight  around  the  rocks. 
Although  much  inferior  in  size  to  the  lammergeyer  of 
the  Alps  ((t.  harhaUis),  being  two  feet  less  in  the  extent 
of  the  wings,  he  is  a  noble  fellow,  the  rich  orange  under- 
parts  and  white  heads  of  the  older  birds  rendering  them 
no  less  conspicuous  than  their  size.  Around  the  camps, 
not  only  at  Senafe,  but  in  many  other  places,  they  were 
very  numerous,  and  by  no  means  timid.  Several  were 
shot  on  the  ground.  Indeed,  the  difference  from  a 
country  like  Western  Europe,  where  all  wild  animals  are 
relentlessly  persecuted,  to  one  of  those  favoured  regions 

^  L.  tigrefuisy  sp.  nov.     See  the  description  hereafter. 


FAUNA  OF  SENAFE^BIRDS.  45 

where  they  are  allowed,  with  but  few  exceptions,  to  live 
unmolested,  is  nowhere  more  conspicuous  than  in  the 
numbers  and  comparative  tameness  of  the  raptorial  birds. 
Bruce  relates  an  instance  of  the  boldness  and  rapacity 
of  a  lammergeyer  in  seizing  meat  actually  being  cooked, 
which  exceeds  anything  I  heard  of. 

I  noticed  no  owls,  parrots,  or  woodpeckers  at  Senaf^, 
and  the  only  barbet  t^as  Pogonorhynchus  undatus.  I 
never  met  with  any  cuckoo,  nightjar,  roller,  or  kingfisher. 
The  Abyssinian  representative  of  the  Alpine  swift  {Cyp- 
selus  CBquatoinalis)  abounded  at  the  back  of  Senaf^  rocks. 
Irissor  erythrorhynchics  was  common  in  small  flocks 
on  Akiib  Teriki.  Near  Senaf^  I  first  met  with  the  great 
Abba  Gamba  {Bucorvus  ahyssinicus),  that  most  astounding 
of  birds,  belonging  to  a  gioup  exceptionally  fi:amed  for 
a  life  in  trees,  and  in  general  purely  frugivorous,  but 
itself  dwelling  on  the  ground  and  feeding  on  insects. 
Tockus  Hemprichii  {T.  limbatus  of  Euppell)  also  oc- 
curred, replacing  T.  Jlavirostris  of  the  passes,  which  at 
this  season  did  not  ascend  so  high.  Three  sun-birds  were 
met  with,  Nectarinia  ajffinis,  the  superb  N.  Tacazze,  and 
N.  cruentata,  the  last  rare. 

Lanius  huTneralis  was  the  common  shrike,  Laniarius 
erythropterus  being  far  from  rare  in  the  bushes.  I  occa- 
sionally saw  that  curious  dull-coloured  flycatcher  Brady- 
omis  chocolatinay  besides  the  little  Platysteira  pririt. 
The  Abyssinian  thiiish  {Turdus  simeims)  abounded,  and 
T.  olivacinus,  a  bulbul  {Pycnonotus  arsinoe),  Crateropus 
leucopygiuSy  and  several  Di^mcecce  were  met  with. 
Amongst  the   chats,  Saxicola  hgens  and  S.  isahellina 


46  PERSONAL  NARRATIVE. 

were  exceedingly  numerous ;  S.  cenanthe  was  less  com- 
mon ;  &  lugubris  kept  much  to  rocky  places ;  ITiamr 
nobia  melcBna  was  not  rare  amongst  bushes  on  hill-sides. 
Of  pipits,  AnthiLS  sordidus  was  met  with  on  barren  slopes, 
A.  ceruinus  in  ploughed  fields,  and  A.  campestris  in 
meadows ;  wliilst  only  two  larks  occurred,  both  apparently 
undescribed  species,  a  Cahndrella  and  an  Alauda.  The 
only  crow  was  Corviis  affinis.  The  starling  group  was 
represented  by  two  species  of  Amydrus  {A.  Ruppelli  and 
A.  alhirostns\  and  by  Buphaga  erythrorhyncha,  there 
being  at  this  season  no  Lamprotornis  on  this  part  of 
the  highlands,  although  later  L.  chalyhcem  abounded. 
Hyphantornis  Guerini  in  winter  dress,  Estrelda  minimxi^ 
E.  phcBnicotLH,  Crithagra  striolata,  and  the  common 
Abyssinian  house-sparrow  {Passer  Swainsoni)  were  the 
principal  finches,  though  a  few  rare  species  were  also 
obtained. 

Pigeons  and  doves  were  numerous ;  Coluinba  guinea^ 
C,  alhitorques,  and  Turtur  lugens  all  being  common. 
The  guinea-fowl,  Numida  ptilorhyncha,  was  just  as 
abundant  as  near  the  coast  Two  francolins,  F.  Erkelii 
and  F.  gutturalis^  were  common,  and  F.  Ruppelli  was 
shot,  but  at  a  somewhat  lower  level  only.  A  large  bus- 
tard, doubtless  Otis  Rhaad^  was  once  killed,  but  I  did  not 
myself  meet  with  any  of  the  family.  Of  waders,  Chettusia 
melanoptera  (the  common  Abyssinian  plover)  and  Ibis 
Harpiprion  caruncxdata  were  the  only  species  frequently 
seen;  and  the  Egyptian  goose,  Ch^nalopex  CBgyptiaca, 
was  the  sole  common  representative  of  the  duck  tribe, 
— but  there  was  but  little  wat<jr  in  the  neighbourhood. 


SENAFE^EXCURSION  TO  HAUL  47 

Life  at  Senaf^  was  on  the  whole  very  enjoyable.  A 
perfect  climate,  fine  scenery,  a  novel  fauna,  agreeable 
companions,  and  a  good  commissariat,  combined  to  make 
the  time  pass  pleasantly.  The  whole  country  around 
appeared  perfectly  safe,  and  the  people  thoroughly 
friendly,  so  far  as  I  could  see,  and  they  were  certainly 
an  improvement  on  Shohos  both  in  manners  and  personal 
appearance.  Still  they  are  a  poor  race,  often  half  starved, 
and  very  inferior  to  the  dwellers  in  the  rich  valleys  of 
Central  Abyssinia. 

After  I  had  been  a  few  days  in  Senafe,  Sir  Robert 
Napier's  well-known  order  was  issued,  that  all  servants 
except  grass-cutters  for  the  horses  should  be  dismissed, 
and  only  seventy-five  pounds  of  private  baggage  be  carried 
forward  by  each  officer.  Personally,  of  course,  I  could 
travel  as  lightly  as  anybody  else,  and  if  I  renounced  all 
attempts  at  collecting  any  specimens,  either  geological  or 
zoological,  I  might  have  still  gone  on  and  examined  the 
geology.  But  I  thought  that  after  I  had  succeeded,  with 
great  difficulty,  in  bringing  collectors  and  all  the  neces- 
sary apparatus  to  the  highlands,  it  would  be  a  pity  to 
take  them  no  further  than  SenafS,  and  I  accordingly 
wrote  to  head-quarters,  then  near  Antalo,  and  applied 
for  additional  carriage.  Meantime  Captain  Carter,  the 
energetic  head  of  the  Trigonometrical  Survey,  had 
reached  SenafS,  and  intended  going  to  Takonda  and 
HalaL  He  asked  me  to  accompany  him,  and  I  was  very 
glad  of  the  opportunity  of  seeing  so  classical  a  region  as 
the  head  of  the  Taranta  pass,  traversed  by  aU  the  earlier 
explorers  of  Abyssinia. 


48  PERSONAL  NARRATIFR 

We  were  to  have  started  on  the  26th  February,  but 
a  disturbance  arose  between  some  of  the  Tigr^an  troops 
in  the  neighbouring  villages  and  the  followers  of  a  rebel 
chief,  who  liad  endeavoured,  more  Abyssinico,  to  annex 
territory,  and  it  was  considered  by  Major  Roome,  the 
political  ofl&cer,  not  advisable  that  a  small  party  should 
leave  the  camp  until  matters  were  quieter.  However, 
next  day  everything  was  again  peaceable,  and  we  started. 
The  road  led  over  the  shoulder  of  Akiib  Teriki  by  a  path 
composed  of  huge  blocks  of  stone,  then  along  a  terrace- 
like flat,  and  across  a  saddle  dividing  two  great  valleys, 
one  running  to  the  Komayli  pass,  the  other  into  the 
Hamas  valley,  a  feeder  of  the  Mareb.  The  latter  afforded 
a  superb  view,  being  bordered  wdth  huge  precipices  of 
sandstone  and  trachyte,  the  heights  fringed  with  juniper 
forest.  The  juniper  generally  grows  freely  on  the  sand- 
stone, and  its  deep  sombre  cypress  colour  contrasts 
finely  with  the  white  of  the  rocks. 

I  foimd  Carter,  who  had  started  before  me,  upon  a 
rise  beyond  the  saddle,  hard  at  work  with  a  plane-table 
mapping  the  hills  around.  My  mules  detained  us,  as, 
although  only  very  lightly  laden,  their  loads,  owing  to 
the  badness  of  the  saddles  and  the  want  of  skill  amongst 
the  drivers,  slipped  off*  as  usual  at  every  little  ascent 
or  descent,  so  that  we  did  not  reach  our  camping-ground 
till  dark.  It  was  a  bitterly  cold  night,  and  in  the  morn- 
ing the  little  meadow  on  which  our  camp  stood  was  white 
with  hoar-frost. 

The  next  day  was  spent  on  the  great  sandstone  plateau 
between  the  Komayli  and  Haddas  valleys.      This  is,  like 


UALAI.  49 

Soweta,  uninhabited  except  in  the  rains,  when  Shohos 
drive  their  cattle  here  for  pasture.  Pieces  of  Greek 
ruins  lie  scattered  about  here  and  there.  We  could, 
however,  only  find  fragments  of  inscriptions,  containing 
two  or  three  letters  at  the  most.  There  was  no  such 
view  now  over  the  deep  valleys  as  when  I  ascended  the 
sides  of  the  same  plateau  fi'om  UnduL  All  below  our 
feet  was  a  dense  sea  -  of  mist,  from  above  which  the 
higher  hills  rose  like  islands  and  promontories.  The 
spring  rains  which  fertilize  the  coast  of  the  Red  Sea  still 
prevailed  in  the  lowlands ;  they  however  do  not  extend 
to  the  Abyssinian  plateau,  and  the  rains  of  summer  are 
similarly  confined  to  the  highlands. 

In  the  evening  we  saw  several  hyaenas,  but  I  could 
not  succeed  in  shooting  any.  The  next  day  we  went  to 
Halai,  over  a  series  of  low  ridges  and  valleys,  with  a  few 
bad  descents.  It  was  a  rather  long  march,  and  we  did 
not  ourselves  arrive  before  nightfall,  having  delayed 
much  by  the  way.  At  one  village  on  the  road  we  found 
a  Catholic  church,  with  a  number  of  the  highly-coloured 
pictures  of  saints  and  of  Biblical  incidents  so  common  in 
smaller  churches  in  France  and  Italy,  but  also  with  some 
of  the  mural  paintings  generally  found  in  Abyssinian 
churches,  St.  George  and  the  Dragon  being  conspicuous 
as  usual.  ^ 

Halai  is  a  considerable  village,  buUt  of  flat-roofed 
houses,  like  all  others  in  this  part  of  Tigr^,  and  inhabited 
by  Christians.      There   is  a  good-sized   square   churcli, 

^  It  is  scarcely  necessary  to  remind  the  reader  that  St  George  is  as  muc 
the  patron  saint  of  Abyssinia  as  he  is  of  Engknd. 

E 


50  PERSONAL  NARRATIVE, 

The  head-man  was  very  civil,  and  sent  us  some  supplit»s. 
We  spent  a  couple  of  days  here,  Carter  mapping,  I 
working  at  the  geology  and  collecting.  The  village  is 
near  the  head  of  the  Taranta^  which,  if  the  information 
obtained  by  Carter's  interpreter  was  correct,  is  not  the 
name  of  a  mountain,  but  of  the  well-known  steep  pass 
leading  from  the  Haddas  valley  to  Dixa. 

The  whole  plateau  around  Halai  is  composed  of  meta- 
morphic  rocks,  as  is  the  northern  part  of  the  great  Tsaro 
spur  to  the  east  of  the  Haddas  gorge ;  sandstone  comes 
in  south  of  Halai,  near  the  village  of  Dhera,  and  south  of 


AtMWAT  piurresu,  E. 

VI.MALAI  PLATEAU, 


T^fW 


^^mM 


this,  as  far  as  Takonda,  caps  all  the  higher  portions  of 
the  plateau,  the  intervening  valleys  being  cut  into  the 
subjacent  metamorphic  rocks.  The  southern  portion  of 
the  Tsaro  or  Asawat  plateau,  east  of  the  Haddas,  is  an 
unbroken  mass  of  sandstone  like  Sowera.  The  greater 
portion  of  the  sandstone  is  white  and  massive,  bands  of 
conglomerate  occasionally  occurring,  sometimes  with  a 
brick-red  matrix,  ferruginous  bands  being  common. 

In  the  Haddas  gorge  a  ridge  capped  with  about  300 
feet  of  sandstone  occurs  in  a  peculiar  manner.  It  is  at  a 
much  lower  elevation  than  the  same  rock  on  the  plateaux 
on  each  side,  its  top  being  about  1,500  feet  below  their 


SANDSTONE  NEAR  HALAI  AND  TAKONDA,  5 1 

level.  It  divides  two  streams :  one  the  main  source  of 
the  Haddas^  running  from  close  to  Takonda ;  the  other  a 
side  valley,  up  which  is  the  pass  to  Takonda  from  the 
Haddas.  This  sandstone  cap,  the  top  of  which  is  far 
below  the  base  of  the  sandstone  on  the  hills  at  both 
side^,  extends  to  the  north  bcj^ond  the  termination  of  the 
sandstone  on  the  plateaux.  It  is  thus  clear  that  the 
somewhat  abrupt  northern  termination  of  the  sandstone 
on  the  plateaux  is  not  due  to  faulting,  but  the  intervening 
ridge  has  much  the  appearance  of  being  let  down  by  two 
great  faults,  one  along  each  of  the  valleys  at  it«  sides. 
It  may  however  owe  its  existence  to  the  sandstone 
being  deposited  on  a  very  uneven  surface,  and  filling 
a  pre-existing  hollow  in  the  metamorphic  rocks  at 
this  point 

There  is,  however,  another  instance  south  of  Takonda, 
in  which  the  sandstones  appear  to  be  thrown  by  a  fault. 
On  the  hill,  just  south  of  the  village,  there  is  a  thin  cap 
of  sandstone  resting  on  metamorphic  rocks.  Immediately 
south  of  this  is  a  deep  valley,  on  the  opposite  side  of 
which  the  sandstones  appear  in  a  narrow  terrace,  about 
500  feet  lower  than  their  position  on  the  north  side, 
although  they  are  horizontal  in  both  places.  The  isolated 
block,  already  mentioned  as  occurring  to  the  west  of  the 
road  from  SenafS  to  Rahaguddy,  is  a  third  case.  All  of 
these  may  be  due  to  deposition  on  the  irregular  surface 
of  the  crystalline  rocks,  but  in  the  two  former  instances, 
at  least,  faulting  appears  more  probable.  As  a  general 
rule,  however,  the  absence  of  all  signs  of  geological  dis- 
turbance  in   the  sandstones   is   remarkable,  since  they 

E  2 


52  PERSONAL  NARRATIVE. 

must  be  of  older  date  than  oolitic,  as  will  be  shown  in  a 
subsequent  page. 

We  marched  back  towards  Takonda  on  the  2d  March, 
as  we  found  that  to  go  on  to  Dixa  involved  a  very  diffi- 
cult descent,  and  we  could  see  the  town  from  near  Halai. 
On  our  return,  we  took  a  more  western  road,  and  came 
to  the  edge  of  the  plateau,  from  which  there  is  a  steep 
descent  to  the  west,  into  the  valleys  around  Dixa. 

Near  Anakonda,  west  of  Takonda,  the  surface  of  the 
sandstone  consists  of  a  hard  white  argillaceous  rock 
From  the  edge  of  the  plateau  in  this  direction  there  is  a 
fine  view,  chiefly  over  a  metamorphic  country,  but  sand- 
stones are  seen  to  the  north-west,  and  to  the  south-west 
are  the  extraordinary  hills  near  Adowa,  said  to  be  of 
trachyte.  In  the  centre  of  the  undulating  plain,  which 
stretches  away  to  the  Adowa  range,  is  a  remarkable 
isolated  pillar-like  hill,  called  Tawhili,  which  lies  about 
west  by  south  from  Takonda,  at  a  distance  of  neai'ly 
twenty  miles,  and,  judging  from  its  form,  is  probably 
also  of  trachyte.  This  is  the  more  interesting  as  it  serves 
to  some  extent  to  indicate  a  connexion  between  the  hills 
of  Adowa  and  the  remarkable  rocky  crags  at  Senaf^. 

We  had  foimd  the  march  from  Takonda  to  Halai  so 
long  and  tedious  that  our  mules  had  not  arrived  till 
nearly  nightfall,  owing,  of  course,  in  great  measure  to 
the  frequent  halts  necessary  for  the  purpose  of  readjust- 
ing loads.  On  our  return  journey  we  halted  half-way,  in 
a  lovely  vaUey,  with  fine  sandstone  clifis  at  the  sides,  a 
running  stream  of  beautifully  clear  cold  water,  a  broad 
expanse  of  turf  in  places,  and,  in  others,  along  the  edge 


HYJBNAS— RETURN  TO  SENAFK  53 

of  the  stream,  some  of  the  finest  jxmiper  trees  I  saw  in 
Abyssinia.  Both  here  and  close  to  Halai  these  trees 
grew  with  a  luxuriance  unusual  elsewhere.  Some  of 
them,  near  our  camp,  served  as  the  roosting-place  for  a 
great  flock  of  guinea-fowls.  As  soon  as  it  was  quite 
dark,  a  couple  of  hyaenas  came  to  drink  at  the  stream 
close  by  our  tent.  The  night  before,  at  Halai,  one  of 
these  brutes  had  seized  a  goat  which  was  tied  up,  and 
broken  his  jaw,  and  after  being  driven  off  had  returned, 
and  severely  wounded  a  pony,  tearing  open  his  thigh. 
The  boldness  of  the  Abyssinian  hyaenas  is  remarkable ; 
they  appear  to  think  nothing  of  tearing  or  attempting  to 
tear  pieces  of  flesh  from  the  sides  of  living  animals  as 
large  as  a  mule  or  pony.  On  this  occasion,  however,  it 
being  bright  moonlight,  I  went  after  one  fellow,  and 
by  good  luck  put  a  bullet  through  his  neck,  dropping 
him  on  the  spot.  It  is  always  more  by  accident  than 
anything  else  that  one  succeeds  in  killing  an  animal  by 
moonlight.  No  plan  I  ever  tried — and  I  have  tried 
many — really  enables  a  sportsman  to  see  along  the 
barrels  of  his  gun,  except  in  the  unusual  circumstance 
of  the  moon  being  in  such  a  position  as  to  be  reflected 
from  the  rib  between  the  barrels.  This  once  happened  to 
me  in  India :  I  went  after  a  bear,  who  came  close  to  my 
tent  at  night,  and  catching  by  chance  the  moonlight  on 
my  barrels,  I  shot  him  through  the  heart  with  as  much 
certainty  as  in  daylight. 

The  next  day  we  marched  to  Takonda  again,  and 
thence  into  SenafS.  We  had  run  out  of  commissariat 
supplies,  Ijut'  guinea-fowls,  hares,  and   partridges  were 


54  PERSONAL  NARRATIVE. 

abundant^  and  we  shot  sufficient  not  only  for  ourselves, 
but  to  a  great  extent  for  our  camp. 

At  Senaf^  I  received  an  answer  from  Colonel  Dillon, 
the  military  secretary  to  Sir  R.  Napier,  telling  me  to 
purchase  any  carriage  I  wanted  for  specimens,  as  none 
could  be  supplied  by  the  Transport  Train.  For  some 
days  I  could  obtain  no  mules,  as  none  were  brought  for 
sale.  At  last  I  succeeded  in  obtaining  three.  Drivers 
were  equally  unprocurable,  imless  I  took  Shohos,  who 
were  sure  to  be  missing  after  two  or  three  stages, 
perhaps  taking  the  mules  with  them ;  however,  my 
Indian  horsekeepers  tried  their  hands  at  mule-driving, 
and,  the  little  Abyssinian  mules  being  really  very 
manageable,  they  succeeded  pretty  well. 

It  was  thus  the  1 7th  of  March  before  I  could  start  for 
the  front.  By  this  time  the  head-quarters  were  close  to 
Ashangi,  and,  as  we  then  supposed,  within  two  or  three 
marches  of  Magdala.  But  the  information,  so  far  in  the 
rear  as  Senaf^  then  was,  of  the  progress  of  head-quarters 
was  most  scanty,  and  all  sorts  of  rumours  prevailed. 
Had  the  march  forwards  from  Antalo  been  as  leisurely 
as  that  from  Senaf6  and  Adigrat,  there  would  still  have 
been  abundant  time  for  me  to  reach  the  front  before  the 
army  could  amve  at  its  goal ;  but  lately  the  progress 
had  been  much  more  rapid. 

It  was,  of  course,  of  the  gi-eatest  importance  to  me 
that  I  should  see  as  much  of  the  countiy  as  possible,  in 
order  to  be  able  to  form  a  clear  idea  of  its  geolog}'.  I 
certainly  had  no  idea,  even  at  this  period  of  the  cam- 
paign, that  its  end  was  so  near,  and  I  believe  that  my 


MARCH  FROM  SENAFE  TOWARDS  MAODALA.  55 

expectations  that  a  portion  at  least  of  the  army  would 
remain  in  Abyssinia  throughout  the  monsoon  were  shared 
by  a  large  majority,  if  not  by  the  whole,  of  the  expedi- 
tionary force.  As  will  be  seen,  I  left  just  in  time  to 
catch  the  force  before  Magdala. 

Leaving  Senaf^  the  road  traverses  a  plain  of  meta- 
morphic  rocks,  excessively  slaty  for  the  most  part,  and 
showing  very  little  crystalline  structure.  To  the  west 
is  a  range  of  hills  capped  by  sandstone ;  to  the  east 
the  slaty  and  schistose  formations  form  rounded  masses, 
between  which  flow  the  small  streams  which  form  the 
heads  of  the  valleys  running  towards  the  Salt  Plain  near 
Amphila.  There  is  but  little  of  interest  on  the  road  until 
it  approaches  Guna  Guna,  twelve  miles  from  SenafS. 
Here  the  sandstone  scarp  to  the  west  approaches  nearer 
to  the  road,  and  in  a  small  cleft  in  it  is  the  shrine  of 
St.  Komanos,  where  lie  his  bones  and  those  of  his  com- 
panions. The  spot  is  a  very  lovely  one  :  a  little  stream 
runs  through  a  ravine  with  precipitous  sides,  the  bottom 
filled  with  magnificent  trees.  A  little  farther,  the  road 
enters  the  valley  of  Guna  Guna,  with  its  flat  bed  of 
rich  turf  surrounded  by  magnificent  sandstone  precipices, 
half-way  up  one  of  which  is  a  church  cut  in  the 
rock. 

I  stayed  here  a  day  examining  the  sandstones,  which 
are  precisely  similar  to  those  of  Sowera,  aud  in  which  I 
could  find  no  traces  of  fossils.  I  shot  a  sparrow-hawk, 
which  I  subsequently  found  to  be  a  fine  old  specimen 
of  the  rare  Accipiter  imduliventer  of  RUppell,  and 
I  also,  for  the  first  time,  saw  that  remarkable  African 


56  PERSONAL  NARRAIWR 

bird,  the  coly.  The  species  was  Colitis  leucotis,  peculiar 
to  the  Abyssinian  highlands.  They  were  in  small  flocks 
in  thick  busL  On  the  sandstone  cliflfs  a  large  finely- 
coloured  lizard  (SteUio  cyanogaster)  alx)unded.  In  order 
to  obtain  specimens  I  was  compelled  to  shoot  them,  as 
otherwise  they  were  soon  beyond  my  reach. 

I  had  to  return  to  SenafS  for  a  day  to  make  some 
arrangements,  and  I  did  not  finally  leave  Guna  Guna 
till  the  21st.  The  3d  Dragoons  were  also  marching  to 
the  front,  and  their  baggage  and  commissariat  rather 
encumbered  the  narrow  road.  Ccaptain  Roddy  of  the 
Punjab  Transport  Train  was  also  passing  through  in 
charge  of  a  detachment  of  mules  loaded  with  spare 
ammunition.  I  rode  on  with  Captain  Newport  of  the 
Commissariat,  who  was  on  his  way  to  take  charge  of  the 
station  at  Adigrat 

From  the  valley  of  Guna  Guna  the  road,  inclining 
slightly  to  the  west,  rises  by  a  steep  ascent  to  the  sand- 
stone plateau,  here  very  narrow,  and  forming  the  divid- 
ing ridge  between  the  streams  running  to  the  NUe 
valley  and  those  flowing  into  the  Salt  Plain.  Right  and 
left  of  the  road  deep  ravines  arc  cut  through  the  sand- 
stones into  the  metamorphic  rocks.  The  scenery  is  very 
grand,  and  increases  in  beauty  near  Fokada.  Here 
several  hills  of  basalt  and  trachyte  rest  upon  the  sand- 
stone, the  former  being  bedded.  Some  small  hills  west 
of  the  road  are  distinctly  terraced,  and  although  of 
but  small  size  they  appear  to  consist  of  two  or  three 
beds.  Close  to  Fokada  there  is  a  fine  hill  of  columnar 
trachyte    cast  of  the   road ;   the   rock    of  which   it   is 


FOKJDA.  5  7 

composed  closely  resembles  that  of  Kishyat,  being  finely 
crystalline^  and  grey  or  pale  lilac  in  colour.  It  is  very 
probably  a  portion  of  the  same  great  flow  as  Kishyat; 
but  the  trachyte  at  Fokada  unmistakeably  rests  upon 
basalt,  which  again  overlies  sandstone.  The  basalt  is 
minutely  crystalline,  and  dark  greyish-green  in  colour. 
This  hill  of  Fokada  tends  very  strongly  to  connect  the 
traps  of  Senafe  with  those  to  tho  south  near  Adigrat. 

Where  the  road  winds  round  the  western  side  of  the 
great  trachyte  hill,  the  view  over  the  valleys  to  the  west- 
ward is  one  of  unusual  interest  and  beauty  (see  Plate  III.). 
The  valleys,  as  usual,  are  deeply  cut  into  the  metamor- 
phics ;  the  flat  hill-tops  are  of  sandstone.  To  the  south- 
ward, above  the  sandstone-bed,  rise  the  terraced  trap 
hills  of  the  Harat  range.  In  the  far  distance  are  the 
strangely-shaped  hummbcks  of  the  AJowa  mountains. 

While  traversing  this  part  of  the  road  1  was  suddenly 
surprised  by  hearing  the  most  hoarse  and  guttural  of  croaks 
that  ever  issued  from  a  bird's  throat,  and  looking  upwards 
saw  two  or  three  of  the  remarkable  thiek-l)illed  caiTion 
crow  {Cormdtur  crassirostris)  sitting  with  vultures  on  a 
tree,  over  the  carcase  of  a  camel.  This  curious  bird  was 
comparatively  rare  and  local  in  this  part  of  Tigrd,  but  I 
subsequently  saw  it  abundant  on  the  high  plateaux  of 
Lasta.  It  has  a  singular  and  by  no  means  prepossessing 
appearance,  with  its  disproportionate  bill  and  the  curious 
white  patch  at  the  back  of  the  head. 

Close  to  Fokada  I  also  first  met  with  that  lovely  re- 
presentative of  the  starling  group,  LamprocoUus  chahj- 
h(mi9y  which,  later  in  the  year,  abounded  in  this  portion 


58  PERSONAL  NARRATIVE. 

of  Tigr^.  Every  day,  indeed,  as  we  marched  through 
these  beautiful  highlands,  new  kinds  of  birds  and  animals 
were  noticed.  But  little  could  be  collected,  as  there  was 
no  time  to  hunt  for  specimena 

The  camp  at  Fokada  was  beautifully  situated  at  the 
south  side  of  the  great  trachyte  hill,  and  at  the  head  of 
a  deep  ravine  in  the  sandstone  leading  to  the  eastward, 
the  resort  of  many  lammergeyers,  eagles,  vultures,  and 
kitea  Coi^ultur  was  also  common.  On  the  next  day  I 
marched  to  Adigrat.  The  road  lies  along  the  eastern 
scarp  of  the  Harat  range,  and  is,  for  the  greater  part  of 
the  distance,  upon  basalt,  the  band  of  sandstone  being  at 
a  lower  level ;  but  there  is  a  descent  to  the  sandstone 
again  before  reaching  Adigrat.  To  the  east  are  the 
usual  deep  valleys  between  hills  capped  with  sandst^one  ; 
to  the  west  rise  the  massive  spurs  of  the  great  Harat 
range. 

Time  did  not  allow  of  my  exploring  the  Harat  hills, 
either  when  marching  to  Magdala  or  when  returning,  and 
I  could  only  examine  the  lower  portions.  These  con- 
sist of  basalt  alone,  frequently  amygdaloidal,  in  well- 
marked  horizontal  beds,  the  terracing  so  characteristic  of 
trap  rocks  being  most  strongly  marked,  even  more  so 
than  in  the  trap  ranges  of  Western  India.  From  my 
subsequent  experience  of  the  ranges  in  Lasta,  I  have  but 
little  doubt  that  the  upper  portions  of  the  Harat  hills  are 
largely  composed  of  trachyte  flows,  and  many  of  these 
appear  to  be  of  gi*eat  thickness.  One,  forming  a  cap 
with  perpendicular  sides  on  the  top  of  the  two  highest 
peaks  of  the  whole  range, — the  anibas,  or  hill  forts,  of 


AhlGRAT.  59 

Alekwa  and  Andale, — looks  from  a  distance  as  if  it  could 
not  be  less  than  200  feet  thick ;  probably  it  is  more. 

Adigrat  itself  stands  on  the  sandstone  terrace  at  the 
base  of  the  traps.  It  is  a  considerable  town,  with  a  fine 
chmxjh,  contaming  some  remarkable  mural  paintings,  in 
which  Scriptural  scenes  are  portrayed,  as  they  might 
have  appeared,  perhaps,  had  the  scene  been  Abyssinia 
and  the  actors  Abyssinians,  just  as  the  Italian  painters  of 
the  Middle  Ages  introduced  the  costumes  of  Italy  and 
the  great  buildings  of  Florence  and  Sienna  in  the  repre- 
sentation of  events  which  occurred  in  Palestine. 

On  the  road,  a  few  miles  north  of  Adigrat,  numerous 
large  rolled  blocks  of  granite  lie  scattered  about,  appa- 
rently derived  from  the  sandstone.  This  is  the  only 
instance  I  noticed  of  the  occurrence  of  rounded  granitic 
blocks  on  the  table-land. 

As  far  as  Fokada,  all  the  Abyssinian  dwelling-places 
seen  had  been  square,  with  flat  roofs  like  those  at  Senafd, 
and  usually  built  against  the  slope  of  a  hill ;  but  about 
Adigrat  we  came  upon  the  round  huts  with  conicid 
thatched  roofs,  which  appear  to  be  the  prevalent  form 
throughout  Central  Abyssinia,  for  no  great  change  took 
place  between  this  and  Magdala.  The  difierence  may  be 
paitly  due  to  the  heavier  rains  further  south — mainly, 
however,  I  was  told,  to  the  paucity  of  grass  for  thatching 
in  Agame.  The  round  thatched  huts  are  certainly  far 
better  adapted  to  keep  out  rain. 

At  Adigrat  I  left  a  collector  and  a  peon,  with  instruc- 
tions to  obtain  for  me  skins  of  as  many  mammals  and 
large   birds   as   they   possibly   could,   Captain   Newport 


60  PERSONAL  NARRATIFE. 

kindly  taking  charge  of  the  men.  1  diminished  my  own 
kit  as  much  as  I  could,  and  re-sterted  the  followmg 
morning,  taking  with  me  only  one  of  my  native  collectors, 
a  very  good  skinner,  but  lame,  so  that  I  had  to  carry  him 
on  my  spare  horse  as  long  as  I  had  one. 

From  Adigrat  to  the  next  camping-ground  to  the 
south,  Mai  Wahiz,  the  road  runs  upon  the  sandstone 
terrace,  and  for  the  first  twelve  miles  it  continues  to  skirt 
the  great  range  already  described.  This  range  terminates 
to  the  south  in  some  noble  spurs,  with  very  precipitous 
sides  of  horizontal  trap,  resting  upon  sandstone ;  and  the 
road  then  emerges  into  a  more  level  country  than  any 
previously  traversed.  The  sandstone  belt  is  continuous 
all  the  way  from  Senafd,  and  nearly  horizontal  through- 
out, although  never  apparently  much  above  800  to  1,000 
feet  in  thickness. 

Between  Adigrat  and  Mai  Wahiz  is  the  highest  part 
of  the  route  in  Tigrd,  the  road  being  at  one  spot,  about 
half-way  between  the  two  camps,  upwards  of  9,000  feet 
above  the  sea.  I  here  first  met  with  a  little  blac*k  and 
white  Pratincolay  much  resembling  the  common  P.  ca- 
prata  of  India,  but  with  more  white  about  the  male,  and 
a  more  rufous  female.  This  bird,  although  far  from 
scarce,  appears  to  have  escaped  the  notice  of  all  previous 
naturalists  in  Abyssinia,  until  described  by  Von  Heuglin 
in  1869  as  P.  semitorquata.  I  also  here  first  saw  and 
collected  the  beautiful  little  green  parroquet  Psittacttla 
TarantcB,  found  by  Mr.  Salt,  and  described  in  the 
Appendix  to  his  "Journey." 

Hitherto,   from   Senafe,  the   road,  where   it  did  not 


MAI  WAHIZ,  61 

actually  run  along  the  crest  of  the  dividing  range,  kept 
slightly  to  the  east  of  it ;  but  from  Mai  Wahiz,  nearly  to 
Lake  Ashangi,  aU  the  small  streams  crossed  on  the 
route  run  to  the  westward  into  the  Nile  basin.  All  are 
small,  the  dividing  ridge  between  the  two  watersheds 
being  only  a  few  miles  distant  during  the  whole  route 
until  after  passing  the  Takkazzyd. 

From  the  plain  near  Mai  Wahiz,  the  Haramat  ranges 
near  Adowa,  so  frequently  mentioned  before  in  con- 
nexion with  the  Senaf^  rocks,  are  very  conspicuous,  at 
no  great  distance  to  the  west. 

Just  south  of  the  camping-ground  at  Mai  Wahiz,  a 
conspicuous  basalt  dyke  crosses  the  road.  It  is  curved, 
but  has  a  general  north-west  strike.  Trap  dykes,  as  a 
rule,  however,  are  rare  throughout  the  country  traversed. 
Some  fragments  of  iron  ore,  evidently  containing  man- 
ganese and  resembling  psilomelane,  were  here  met 
with,  evidently  derived  from  the  sandstone,  \>emg  only 
a  peculiar  form  of  the  ferruginous  bands  which  occur 
80  frequently  in  it. 

About  four  or  five  miles  south  of  Mai  Wahiz  there  is 
a  small  sandstone  scarp,  down  which  the  road  descends 
to  a  lower  plain,  chiefly  consisting  of  metamorphic  rocks, 
of  which  a  large  expanse  to  the  westward  is  composed, 
while  to  the  eastward  the  sandstone  range  continues 
to  the  parallel  route.  The  road  traverses  metamorphics 
till  past  Adabdgi,  the  next  camping-ground,  and  nearly 
as  far  as  Dongolo.  The  general  strike  of  their  foliation 
is  north  and  south,  as  near  Senafe,  varying  about  10**  to 
the  east  or  west     South  of  Adabdgi  the  sandstone  range 


62  PERSONAL  NARRATrrR 

falls  back  to  the  eastward,  leaving  a  wider  plain  of  meta- 
morphic  rocks  in  that  direction. 

At  AdaMgi  my  progress  was  very  nearly  brought  to  a 
stop  by  all  my  mules  knocking  up.  The  Otago  saddles 
which  I  had,  though  fairly  adapted  for  horses  or  very 
large  mules,  were  quite  unsuited  to  the  little  Aljyssinian 
animals,  and  galled  their  shoulders  and  hips  fearfully. 
Hundreds,  if  not  thousands,  of  mules  were  utterly  ruined 
by  these  saddles  during  the  campaign.  One  animal  of 
mine  was  so  lame  from  a  shoulder-gall  that  I  was  obliged 
to  leave  him  behind,  and  the  others  were  unable  to  carry 
their  loads  for  a  day  or  two.  In  this  emergency  I 
loaded  my  horses,  and  Captain  Roddy,  having  some  spare 
animals  in  his  train,  very  kindly  carried  a  load  for  me 
for  a  day  or  two  until  my  mules  recovered.  I  subse- 
quently procured  Punjab  packs,  consisting  of  two 
cushions,  united  by  a  band  over  the  back,  and  on  which 
the  loads,  first  packed  in  canvas  bags,  called  saJetas  in 
India,  are  lashed  with  a  rope  ;  and  with  these  the  mules 
got  on  well  enough.  Unfortunately  the  loads  had  a 
great  tendency  to  shift,  and  required  constant  adjusting 
on  the  road,  but  my  men  soon  became  fairly  expert  at 
mule-loading.^ 

^  I  believe  that  eveiy  one  who  had  experience  in  miile-carriage  utterly 
condemned  all  the  variouB  saddles  used  in  the  Abyssinian  expedition.  My 
own  experience  was  the  following  : — 

The  **  Otago  "  saddle,  though  heavy  and  cumbersome,  is  suitable  for  roads 
without  many  steep  ascents  and  descents,  provided  it  be  fitted  to  the  animal 
carefully;  those  sent  to  Abyssinia  were,  as  a  rule,  too  large  for  mules. 
On  steep  roads,  from  being  hard  and  stiff,  it  presses  too  much  against;  the 
shoulders  and  hips.  The  Bombay  pad  was  simply  ridiculous.  The  Punjab  pad 
was  by  far  the  best  form  used,  for  small  animals  especiiilly,  but  it  was  difficult 


LONGOLO,  63 

Leaving  Adabdgi,  the  road  for  some  miles  traverses 
the  undulating  metamorphic  tract.  To  the  west,  a  few- 
miles  distant^  are  some  massive  isolated  crags,  evidently 
of  sandstone,  no  longer  horizontal,  but  tilted  up  at  a  con- 
siderable angle.  About  eight  miles  from  the  camping- 
ground  a  steep  scarp  is  met  with,  down  which  the  road 
descends  to  the  valley  of  the  Genfel.  The  whole  descent 
is  over  metamorphic  rocks,  but  just  beyond  the  base 
sandstone  comes  in,  resting  upon  the  metamorphics,  and 
dipping  at  a  sharp  angle,  about  25%  to  the  south-west. 
This  sandstone  is  evidently  a  continuation  of  the  crags, 
composed  of  similarly-inclined  beds,  already  seen  to  the 
westward  from  the  top  of  the  descent. 

The  valley  now  reached,  in  which  stood  the  camping- 
ground  of  Dongolo,  oflFered  a  great  contrast  to  the 
somewhat  barren  table-land  from  which  we  had  just 
descended.  A  small  stream  runs  thi'ough  the  bottom, 
which  is  richly  covered  with  trees  and  bushes ;  and  the 
massive  bluffs  of  sandstone,  of  a  rich  red  colour,  not 
white,  as  farther  north,  almost  recall  some  of  the  scenery 
in  the  old  red  sandstone  of  Western  England.     Despite 

to  strap  the  load  firmly  if  the  shape  of  the  latter  was  at  all  inconvenient. 
That  any  saccess  was  achieved  by  the  Transport  Corps  was  due  to  the  sheer 
hard  work  of  the  officers,  and  especially  of  those  on  the  highlands.  The 
muleteers,  as  a  rule,  were  abominable.  The  Punjabees  were  manageable,  but 
knew  nothing  of  mules  ;  the  Egyptians  knew  something  of  mules,  but  were 
difficult  to  manage ;  and  the  Persians,  by  far  the  best  muleteers  in  the  army, 
were  few  in  nimiber,  and  disgusted  at  finding  no  one  who  could  understand 
them.  Indeed,  nearly  the  whole  difference  in  respect  to  management  between 
the  unruly  Turks,  Egyptians,  and  Persians,  and  the  easily-managed  Punjabees, 
lay  in  the  simple  fact,  that  nearly  all  the  officers  of  the  Transport  Train  spoke 
Hindustani  fluently,  and  could  converse  with  the  latter,  whilst,  not  under- 
standing Arabic,  they  could  not  make  themselves  intelligible  to  the  former. 


64  PERSONAL  NARRATIVE. 

the  diflfcrencc  in  colour,  I  consider  this  sandstone  iden- 
tical with  that  left  behind  at  Adabdgi,  but  the  question 
will  be  discussed  in  a  subsequent  page  especially  devoted 
to  geology.  In  this  valley  the  beautiful  iridescent  star- 
ling {Lamprocoliiis  chalyhceus)  abounded,  and  many 
birds  reappeared  which  had  been  wanting  on  the  bleak 
plateau  we  had  tmversed  from  Adigrat.  From  this  point 
the  general  elevation  of  the  road  as  far  as  Antalo  did 
not  much  exceed  7,000  feet  above  the  sea. 

From  Dongolo  we  marched  the  same  afternoon  to 
Agula,  the  next  camping-ground.  The  road  passes 
through  a  beautiful  little  glen  in  the  sandstone,  at  one 
side  of  which  is  a  church  cut  in  the  rock.  Immediately 
on  leaving  this,  the  sandstone  distinctly  dips  under 
limestone,  and  the  road  enters  a  large  open  plain  with 
hills  to  the  eastward.  It  traverses  the  level  ground  for 
four  or  five  miles,  and  then  ascends  a  somewhat  barren 
rise,  still  keeping  on  limestone,  which  indeed  wjis  tra- 
versed from  this  point  for  upwards  of  seventy  miles,  and 
until  we  had  passed  Antalo.  In  this  limestone  I  at  last 
found,  fossils,  but  they  were  extremely  obscure.  A  few 
oysters  of  the  Exogyra  type,  and  some  spines  of  Echino- 
derms,  were  the  only  remains  I  could  recognise.  Casts 
of  bivalves  were  common  enough,  but  they,  of  course, 
were  useless  for  determination. 

Not  only  was  the  geology  of  much  interest,  but  several 
birds  which  I  had  not  previously  seen  attracted  atten- 
tion. I  shot  a  pair  of  a  handsome  sand-grouse,  which  I 
afterwards  found  to  be  a  Cape  species,  Pterocles  guttu- 
rails,  and   a  specimen   of  the   long-tailed  dove,   GKna 


AGULA.  65 

capensis.  This  little  dove,  which  in  May  and  later 
abounded  on  the  sea-coast,  appeared  in  winter  and  spring 
to  be  entirely  wanting,  not  only  on  the  shores  of  the  Red 
Sea,  but  throughout  the  portion  of  Tigr^  which  I  tra- 
versed north  of  the  Genfel  river. 

Between  hunting  for  fossils  and  looking  after  birds 
time  slipped  away.  I  had  already  been  detained  in  con- 
sequence of  one  of  my  men  having  been  badly  kicked  by 
a  mule,  and  I  had  put  my  lame  collector  on  a  horse 
which  Captain  Roddy  had  lent  me  to  ride,  and  sent  him 
on ;  so  when  night  came  on,  I  was  alone,  and  still  at  some 
distance  from  camp,  and  I  had  no  Uttle  difficulty  in 
finding  my  way  through  the  darkness.  The  road  was  far 
from  good,  and  after  blundering  over  rocks  for  half  an 
hour,  I  was  glad  to  see  the  camp-fires  of  Agula.  An 
excellent  plan  had  been  adopted  of  pitching  a  large  tent 
— large,  that  is,  for  Abyssinia ;  in  India  we  should  have 
thought  it  very  small — at  each  halting-place,  for  the  use 
of  officers  marching  through.  I  had  scarcely  reached 
this  refuge  when  a  downpour  of  rain  accompanied  by 
heavy  thunder  and  lightning  burst  upon  us,  and  lasted 
for  at  least  an  hour.  Hitherto  the  weather  had  been 
lovely,  but  afterwards  showers  fell  not  unfrequently. 

Near  Agula  are  the  ruins  of  a  church,^  described  by  a 
correspondent  of  the  Illustrated  London  News  as  a 
Greek  temple.  The  scenery  around  is  very  character- 
istic of  this  part  of  the  country,  a  comparatively  open 
valley  with  very  little  vegetation  on  the  slopes,  but  a 

^  A  description  with  a  plan  will  be  found  in  Markham's  *'  Abyssinian 
Expedition,"  p.  236» 

F 


66  PERSONAL  NARRATIVE, 

fringe  of  green  bushes  along  the  banks  of  the  stream. 
The  hills  on  each  side  of  the  valley  ascend  in  a  series  of 
small  limestone  scarps. 

Leaving  Agnla  next  morning,  March  26th,  the  road 
led  over  long  ascents  and  descents  through  an  undulating 
country  almost  entirely  composed  of  limestone.  The 
rock  is  in  comparatively  thin  beds,  much  resembling 
lias,  but  lighter  in  colour.  The  high  dip  seen  near 
Dongolo  disappeai-s  immediately  beyond,  and  the  rock  is 
horizontal  or  undulating,  without  any  steady  dip  in  any 
direction.  The  country  is  covered  with  thin  jungle  of 
mimosa  bush.  I  had  no  better  success  than  the  day 
before  in  fossil-hunting,  but  I  had  not  much  time  to 
search ;  there  was  a  long  march  of  sixteen  miles  from 
Agula  to  the  next  camping-ground,  Dolo. 

The  limestone  evidently  extends  for  a  long  distance 
east  and  west  of  the  road.  Some  hills  about  eight  or 
ten  miles  east  of  Agula  may  perhaps  be  of  sandstone, 
but  the  crags  of  the  Geralta  range  run  almost  due  west 
from  Dongolo,  and  the  ground  seen  to  the  westward  from 
Agula  must  consist  almost  entirely  of  limestone,  into 
which  the  various  streams  traversed  by  the  route  of  the 
expedition  at  Dongolo,  Agula,  Dolo,  &c.,  cut  deep  gorges 
before  imiting  to  form  the  Geba  river,  a  great  affluent 
of  the  Takkazzyd.  This  stream  doubtless  exposes  a  most 
interesting  section  of  the  rocks  in  this  country. 

About  eight  miles  from  Agula  the  road  dips  into  a 
small  valley,  where  is  a  halting-place,  called  Mai  Mak- 
dam.  At  this  spot  an  important  pass  from  the  Salt 
Plain  to  the  eastward  joins  the  route  we  followed,  and  we 


LOW.  67 

here  found  several  dealers  with  loads  of  the  oblons: 
blocks  of  salt  used  by  the  Abyssinians  instead  of  small 
change. 

On  reascending  from  the  valley  at  Mai  Makdam,  basalt 
is  seen  in  a  hiU  east  of  the  road.  A  very  much  larger 
mass  occurs  close  to  Dolo,  and  the  greater  part  of  the 
rock  over  which  the  road  descends  into  the  valley  at  the 
camping-ground  consists  of  it.  The  first  knoll  has  the 
appearance  of  resting  on  the  limestone,  but  the  second  is 
of  such  thickness  that  if  regularly  interstratified  it  would 
be  seen  on  the  hills  around.  It  is  probably  an  intrusive 
mass. 

At  Dolo  I  foimd  Lieut.  St.  John  busily  engaged 
in  putting  up  the  telegraph.  He  gave  me  one  or  two 
birds  which  I  had  not  myself  found,  amongst  them  the 
little  Abyssinian  bustard,  or,  as  Indian  sportsmen  very 
naturally  called  it,  floriken  {Otis  melanogaMer).  It  is  a 
little  larger  than  the  common  floriken  of  Southern  India 
(0.  aurita),  but  has  very  similar  habits,  living  amongst 
bushes  and  high  grass,  usually  crouching  and  seeking  to 
hide  when  approached.  As  in  the  Indian  bird,  the  male 
of  this  little  bustard  changes  in  plumage  with  the  time  of 
year,  the  whole  of  the  under-parts  becoming  black  in  the 
breeding  season.  There  are  other  smaU  African  bustards, 
forming  the  genus  Lissotis  of  some  naturalists,  which 
exhibit  the  same  peculiarity,  and  the  close  affinity  be- 
tween these  African  and  the  Indian  birds  is  very  inter- 
esting, for  this  type  of  bustard  does  not  belong  to  the 
desert  fauna,  and  its  existence  in  both  countries  is  one 
instance  amongst  several  showing  that  the  very  marked 

F  2 


68  PERSONAL  NARRATIFE, 

Ethiopian  relations  exhibited  by  a  large  portion  of  the 
fauna  of  British  India  do  not  arise  solely  from  the  exten- 
sion of  the  desert  types  to  the  sandy  plains  of  North- 
Western  Hindustan,  the  Punjab,  and  Sind. 

In  a  ramble  next  morning  I  found  a  small  boss  of 
granite  exposed  in  the  bed  of  the  stream  just  above  Dolo. 
It  is  of  small  size,  extending  apparently  only  a  few  yards  ; 
and  as  no  sandstone  occurs,  it  is  evident  that  the  limestone, 
here  at  least,  reste  immediately  on  metamorphic  rocks. 
Above  and  below  the  stream  has  cut  its  way  through  lime- 
stone, forming  deep  glens  with  cliffs  at  each  side.  In  the 
bushes  beside  the  stream  I  shot  a  coucal  or  crow-pheasant 
{Centropxis  monachtts),  and  St.  John  killed  a  rail 
{Rallus  Rougeti).  We  made  the  march  to  the  next 
camping-ground,  Haikhallat,  near  Chelikot,  together.  It 
was  over  imdulating  limestone  country,  similar  to  that 
already  traversed.  Harriers  {Circus  cineraceiis  and  C. 
Swainsoni)  abounded,  much  as  in  the  Deccan  in  India, 
but  they  were  even  more  numerous.  I  was  desirous 
to  procure  more  specimens  of  Otis  melanogaster,  and 
we  beat  several  patches  of  bush  without  success,  but 
we  started  two  reddish  antelopes,  one  of  which  we 
succeeded  in  rolling  over.  It  was  a  buck  of  the  Scopo- 
phorus  montanus  of  Riippell,  with  short  straight  horns. 
Another  bush  antelope,  which  we  also  saw  this  day  and 
of  which  I  afterwards  killed  specimens,  was  Cephalophiis 
Tnadoqim  of  the  same  naturalist,  a  mouse-coloured  animal 
about  the  size  of  a  gazelle,  or  rather  smaller. 

On  the  limestone  there  were  a  few  land-shells.  A 
snail  closely  allied  to  the  common  South  European  Helix 


ANTALO.  69 

pisana  covered  some  of  the  bushes  and  the  aloe  plants, 
and  I  found  two  or  three  small  species  of  Pupa,  but,  as 
usual,  only  very  few  kinds  occurred. 

We  rode  on  the  next  day,  the  28th  March,  to  Buya, 
near  Antalo,  over  very  similar  country.  Several  patches 
of  trap  occur,  the  mode  of  occurrence  being  in  general 
rather  obscure,  although  some  of  them  are  clearly  inter- 
stratified  in  the  limestone.  No  dykes  were  observed. 
The  town  of  Antalo  lies  three  or  four  miles  west  of  the 
route  followed  by  the  army,  on  the  side  of  a  hill,  a 
portion  of  which  evidently  consists  of  sandstone,  and 
which  is  very  probably  capped  by  trap.  The  plain  to 
the  east  of  Antalo,  in  which  stood  the  camp  of  Buya,  is 
very  extensive,  and  mostly  covered  by  black  soil,  resem- 
bling the  **regur"  of  India.  The  camp  was  one  of  the 
most  important  dep6ts  of  the  army,  and  a  considerable 
body  of  men  were  stationed  here,  with  large  commissariat 
and  other  stores.  I  never  could  understand  by  what 
remarkable  Itccus  d  non  lucendo  principle  of  nomen- 
clature this  camp,  like  the  corresponding  station  at 
Adigrat,  was  always  spoken  of  as  entrenched,  because  it 
was  surrounded  by  a  wall  of  loose  stones.  I  do  not  at 
all  wish  to  ridicule  this  apparently  very  trifling  defence ; 
it  was  behind  no  better  that  a  whole  army  was  kept 
at  bay  for  weeks  by  a  few  determined  men  at  Cawnpore, 
in  1857 ;  and  even  the  famous  defences  of  the  Lucknow 
Residency  were  not  much  more  formidable  ;  but  the  name 
appeared  to  me  singularly  inappropriate.  On  the  same 
principle  a  walled  city  should  be  one  surrounded  by  a 
ditch — e.g.  Calcutta. 


70  PERSONAL  NARRATIVE. 

At  Buya  I  heard  that  the  army  was  but  little  beyond 
Lat,  about  six  marches  ahead ;  but  it  was  still  supposed 
to  be  but  two  or  three  marches  from  Magdala.  I  halted 
for  a  day,  buying  a  couple  of  fresh  mules  to  replace  those 
which  had  broken  down.  One  of  the  new  ones  objected 
to  be  saddled.  I  obtained  the  assistance  of  some  Egyi> 
tian  muleteers  from  the  Transport  Train,  and  they  very 
soon  put  a  Punjab  pad  on  by  main  force,  three  or  four 
powerful  men  absolutely  holding  the  mule  down  while 
a  couple  of  others  strapped  on  the  pad. 

The  next  day,  March  30th,  I  re-started,  and  made  a 
double  march  to   Meshek,    thus    catching   up    Captain 
Roddy's  mule-train  once  more.     The  first  part  of  the 
march  to   Musgi   is   over  country   composed   of  lime- 
stone, precisely  similar  to  that  traversed  north  of  Antalo. 
Interstratifications  of  trap  are  very  frequently  seen,  and 
one  in  especial  occurs  close  to  the  camping-ground  at 
Musgi.     A  small  stream  emerges  from  the  little  plain 
in  front  of  the  camp,  through  a  narrow  gorge  in  the 
limestone,  which  is  here  horizontal,  or  nearly  so.     The 
stream,  almost  immediately  after  leaving  the  plain,  cuts 
into  a  bed  of  basalt  underlying  the  limestone,  and  this 
basalt  forms  the  channel  of  the  stream  for  a  considerable 
distance.     It  has  a  very  low  dip,  about  2**  or  3**,  to  the 
west.     The  whole  of  the  beds  are  then  cut  ofi*  by  a  fault 
striking  1 5**  west  of  south,  with  an  upthrow  to  the  west ; 
and  the  same  bed  of  trap,  about  100  feet  in  thickness,  is 
seen  on  the  hill-side,  with  limestone  both  above  and 
below  it. 

I  had  ridden  to  Musgi  with  my  mules;  but  thence 


MESHEK,  71 

I  sent  them  on,  and  remained  some  time  examining  the 
rocks.  When  I  re-started,  I  missed  the  road,  and  soon 
found  myself  wandering  amongst  hiUs.  I  came  to  a 
vill^e,  and  succeeded  somehow  in  explaining  that  I 
wanted  a  guide,  and  moreover  in  inducing  a  man,  for 
a  dollar,  to  show  me  the  way  ;  and  he  took  me  back  into 
the  proper  route  before  nightfall,  at  the  spot  where  it 
entered  a  deep  valley,  with  fine  trees  and  a  running 
stream.  Up  this  I  rode  for  some  miles  in  the  dark,  till 
I  reached  the  camp  at  Meshek,  where  I  found  a  party 
of  firiends  wondering  at  my  absence. 

Meshek  is  a  lovely  spot.  A  small  grass  plain,  with  a 
few  scattered  willow-trees,  looking  as  if  made  for  a 
camping-ground,  occupies  a  rather  wide  portion  of  the 
valley.  Above,  there  is  cultivation,  irrigated  by  the  little 
stream  in  the  bottom  of  the  glen ;  below,  dense  forest. 
The  hills  at  the  side  are  composed  of  sandstone  and  lime- 
stone, capped  by  trap.  The  sedimentary  beds  are  greatly 
disturbed ;  near  the  camp  they  are,  in  places,  even  con- 
torted or  vertical.  In  this  part  of  the  valley  the  beds 
appear  to  be  higher  in  the  series  than  the  mass  of  the 
limestone  about  Antalo,  and  sandstone  and  conglomerate 
prevail  to  a  much  greater  extent  than  elsewhere.  These 
rocks  are  dark  in  colour,  and  far  less  pure  than  the  sand- 
stones of  Adigrat  and  SenafS ;  and  I  have  scarcely  any 
doubt  of  their  belonging  to  a  different  series  altogether. 
The  general  dip  is  west  or  south-west.  I  found  a  few 
very  ill-preserved  fossils  in  one  place. 

We  had  now  left  the  comparatively  open  plains  of 
Tigrd,  and  entered  the  deep  valleys  between  the  rugged 


72  'FERSONAL  NARRATIVE. 

spurs  of  the  Lasta  mountains.  The  road  had  been 
much  less  carefully  made  south  of  Antalo,  and  in  many 
places  it  had  suflfered  from  the  traffic,  and  was  but  little 
better  than  any  ordinary  Abyssinian  mule-track.  Leaving 
Meshek,  the  route  led  for  two  or  three  miles  through 
the  bottom  of  the  little  valley,  now  brilliantly  green 
with  the  young  wheat,  and  then,  after  passing  a  large 
roimd  church,  with  a  conical  thatched  roof,  one  of  the 
finest  met  with  on  our  march,  it  commenced  a  steep 
ascent  over  trap  rocks,  which  form  all  the  upper  portion 
of  the  valley.  There  is  an  appearance  here  of  two  dis- 
tinct series  of  traps,  the  lower  slightly  inclined  and 
resting  immediately  on  the  sedimentary  beds,  the  upper, 
which  forms  the  crest  of  the  pass,  distinctly  horizontal ;  • 
but  the  presence  of  two  series  is  not  so  manifest  as  in  the 
next  valley.  Above  the  crest  of  the  pass,  itself  9,800 
feet  above  the  sea,  towers  the  amba  or  hill  fort  of  Alaji, 
the  stronghold  of  Walda  Yasous,  nearly  1,000  feet  above 
the  road.  The  whole  of  this  amba  is  formed  of  hori- 
zontal beds  of  trachyte  and  basalt 

On  this  pass  I  had  first  occasion  to  notice  the  change 
which  takes  place  in  the  vegetation  of  the  Abyssinian 
mountains  at  altitudes  above  or  about  9,000  feet.  A  gigan- 
tic thistle,  ten  or  twelve  feet  high,  and  a  very  fine  species 
of  heath,  often  large  enough  to  be  called  a  tree,  are  two 
of  the  most  conspicuous  plants  above  this  elevation.  A 
corresponding  alteration  takes  place  in  the  fauna,  but  this 
was  less  conspicuous  in  the  Alaji  pass  than  subsequently 
on  some  ranges  of  greater  height. 

From  this  point  to  Magdala  the  whole  route  is  over 


ALAJI FASS—ATALA.  73 

bedded  volcanic  rocks;  the  subjacent  limestones,  sand- 
stones, and  metamorphic  rocks  never  once  emerging 
from  beneath  the  great  trappean  formations. 

From  the  crest  of  the  Alaji  saddle  there  is  a  fall  of 
about  1,500  feet  to  the  valley  of  Atala.  The  greater 
portion  of  this  descent  is  over  horizontal  beds,  mostly- 
basaltic,  a  few  only  being  trachytic ;  but  at  the  bottom 
of  the  vaUey,  just  east  of  the  ground  occupied  by  the 
camp,  is  a  small  ridge  of  basaltic  beds,  dipping  at  an 
angle  of  35**  or  40**  to  30**  south  of  west  These  evi- 
dently belong  to  a  lower  series,  upon  which  the  horizontal 
beds  rest  imconformably,  and  to  which  the  inclined  beds 
seen  in  the  Meshek  valley  also  belong.  Up  the  Atala 
valley  to  the  eastward  all  the  beds  are  horizontal,  like 
those  forming  the  crests  of  the  hills. 

At  Atala  a  large  body  of  Abyssinians,  men  and  women, 
were  assembled.  The  women,  many  of  whom  were  not 
ill-looking,  though  their  beauty  might  have  been  greatly 
heightened  by  the  use  of  water,  were  all  dressed  in  a  long 
shift  of  coarse  cotton,  and  had  their  hair  closely  plaited. 
The  men  wore  the  usual  toga-like  shama,  and  carried 
spears  and  shields.  All  were  much  excited.  The  women 
were  crowded  together  in  the  centre,  one  or  two  of  the 
eldest  and  slu'Ulest  leading  a  sort  of  song,  to  which  the 
others  kept  time  with  much  howling  and  lamentation. 
The  men,  evidently  much  excited,  stood  around.  On 
inquiry  we  learned  that  this  was  an  Abyssinian  indigna- 
tion meeting.  The  poor  people  had  canied  stores  for  the 
commissariat,  and  had  been  paid,  but  Walda  Yasous 
or  his  followers  had  taken  from  them  a  considerable 


74  PERSONAL  NARRATIVE. 

proportion  of  the  money.  The  women  apparently  were 
endeavoming  to  excite  the  men  to  resistance.  Unfor- 
timately,  in  this,  as  in  many  other  cases,  want  of  know- 
ledge of  the  language  and  of  a  good  interpreter  made  it 
impossible  to  obtain  accurate  information. 

The  narrowness  of  the  path  across  the  passes  rendered 
progress  very  slow  upon  these  marches,  as  the  mules  could, 
of  course,  only  go  singly,  and  one  animal  falling,  or  halted 
for  the  adjustment  of  its  load,  stopped  the  whole  convoy. 
The  dragoons,  with  their  long  train  of  commissariat 
supplies,  had  been  in  the  rear  since  Adigrat,  but  by 
forced  marches  they  also  had  come  up,  and  so  much 
delay  was  caused  by  the  great  length  of  the  united 
convoys  that  our  party,  which  had  reached  Atala  from 
Meshek  early  in  the  day,  went  on  over  the  next  spur  to 
Aiba  in  the  afternoon.  The  Aiba  valley  is  higher  than 
that  of  Atala,  but  otherwise  precisely  similar,  an  open 
grassy  flat  at  the  bottom  with  a  stream  nmning  through 
it,  and  hills  bare  of  trees  at  each  side. 

The  Atala  and  Aiba  valleys  recall  most  forcibly  some 
of  the  scenery  in  the  higher  parts  of  the  western  ghAts 
near  Poonah,  in  the  Bombay  Presidency;  and  the  hill 
forts  of  Alaji  and  Daga,  although  grander  and  higher, 
yet  resemble  to  a  certain  extent  the  old  Maratta  strong- 
holds of  Singhur,  Rajghur,  Toma,  &c.  In  each  case  the 
same  geological  peculiarity,  the  occurrence  of  beds  of 
volcanic  rock,  which  weather  with  a  vertical  scarp,  has 
been  employed  for  purposes  of  defence,  both  by  Hindoos 
and  Abyssinians. 

At  the  bottom  of  the  Aiba  valley,  as  at  Atala,  the 


AIBA-FERRJ  PASS.  75 

lower  series  of  traps  crop  out,  inclined  at  a  considerable 
angle,  and  they  are  traceable  to  some  height  south  of  the 
valley,  the  horizontal  beds  resting  quite  unconformably 
upon  them.  The  two  series  are  so  similar  in  general 
character,  that  in  the  hurry  of  travelling  to  the  front, 
although  I  noted  the  dips,  I  attributed  them  to  local  dis- 
turbance, and  it  was  only  when  returning  that  I  clearly 
distinguished  between  the  two  separate  groups  of  beds. 
Nevertheless  the  evidence,  when  examined,  was  found  to 
be  very  complete. 

At  Aiba  I  first  saw  and  shot  a  dwarf  goose  peculiar 
to  Abyssinia  {Beimicla  cyanoptera  of  Rlippell),  which  I 
subsequently  found  abundantly  on  the  higher  plateaux, 
and  which  appears  only  to  be  found  at  considerable 
elevations.  On  the  passes  around  I  saw  a  bunting 
{Emberiza  ccBsia)  and  a  serin  {Serimis  or  Crith<igra 
nigriceps)  w^hich  I  had  not  previously  met  with. 

I  left  Aiba  on  the  1st  April,  and  ascended  the  long 
slope  leading  to  the  Fen\a  pass.  From  the  top  of  the 
inclined  traps,  about  300  or  400  feet  above  the  valley, 
to  the  crest  of  the  pass,  10,500  feet  above  the  sea,  or 
1,500  above  the  valley,  and  again  from  that  to  the  im- 
mense amba  of  Daga,  at  least  1,500  feet  more,  the  whole 
mass  of  the  mountain  consists  of  massive  horizontal  l)eds 
of  dolerite  and  trachyte.  The  latter  appears  to  prevail, 
especially  in  the  higher  part  of  the  pass,  and  in  the  amba 
above,  and  it  is  conspicuous  by  the  much  more  distinct 
vertical  scarps  which  mark  its  outcrop.  It  is  pale  in 
colour,  very  often  brecciated,  some  beds  being  apparently 
in  great  measure  composed  of  volcanic  scoria  and  ash, 


76  PERSONAL  NARRATIVE. 

but  at  the  same  time  highly  crystalline,  and  frequently 
aboimding  in  beautifully-formed  crystals  of  glassy  felspar 
with  rounded  angles.  These  crystals,  from  their  brilliancy, 
are  very  conspicuous  on  the  broken  surface  of  the  rock. 

The  general  effect  of  the  scenery  is  depicted  in  the 
firontispiece,  a  view  of  the  Daga  amba  from  the  south, 
showing  the  horizontal  beds  of  trachyte. 

At  the  summit  of  the  Ferra  pass  all  the  hills  are 
covered  with  a  peculiar  wiry  grass,  which  appeal's  to  l>e 
peculiar  to  the  trachytic  traps. 

The  path  descends  from  the  saddle  to  the  camping- 
gi'oimd  at  Belago,  situated  at  a  very  considerable  eleva- 
tion. Here  the  inclined  beds  of  trap  reappear,  and  hence 
occupy  all  the  coimtry  till  far  beyond  Lake  Ashangi,  with 
the  exception  of  the  top  of  some  of  the  high  ranges 
north-west  of  the  lake.  From  Belago  to  Makhan  the  road 
for  some  distance  proved  uninteresting.  After  one  steep 
descent  it  passed  along  a  flat  valley  to  Makhan,  the  next 
camping-groimd.  Here  we  were  once  more  on  the 
eastern  watershed,  and  continued  east  of  the  dividing 
ridge  until  after  crossing  the  Wombarat  pass  near  Lat. 

The  pass  between  Makhan  and  Ashangi  is  not  so  high 
as  those  further  north,  and  does  not  appear  to  rise  above 
the  top  of  the  inclined  traps,  or  Ashangi  group,  as  I  pro- 
pose to  call  them.  The  road  passes  through  a  superb 
forest  of  junipers,  in  which  the  plantain-eater  wtis  abun- 
dant. Except  around  Meshek  I  had  not  noticed  this 
beautiful  biixi  lately,  doubtless  in  consequence  of  our  road 
having  traversed  open  country.  Ferns  abound,  and  the 
large  heath  attains  magnificent  proportions,  the  climate 


I  ^ 


i] 


!  1 


\ 


ASIIANOL 


i  i 


on  the  eastern  side  of  the  dividing  range  at  this  elevation 
being  probably  damper  than  on  the  w(?st. 

At  the  crest  of  the  pass  we  came  in  sight  of  the  lovely 
little  Lake  of  Asliangi.  From  the  same  spot  there  is  an 
extensive  view  to  the  eastward  over  the  country  of  the 
A^ibo  Gallas.  The  foi-egi-ound  consists  of  finely- wooded 
hills,  all  apparently  of  inclined  beds  of  trap.  Beyond,  at 
a  distance  of  twelve  or  fifteen  miles,  there  is  a  broad 
valley,  probably  2,500  or  3,000  feet  below  the  pass, 
or  about  6,000  feet  above  the  sea.  Beyond  this,  again, 
are  other  rounded  hills,  which  look  as  if  more  thinly 
wooded.  They  may  perhaps  be  metamoq)hic.  From  the 
simmiit  of  the  pass  the  road  descends  rapidly  to  the  l)eaii- 
tifiil  Ashangi  valley,  over  beds  of  amygdaloidal  basalt. 

I  only  halted  for  a  night  at  Ashangi,  and  as  T  i^assod 
some  days  here  on  my  return  from  Alagdala,  I  abstain 
from  description  of  it  for  the  present. 

At  Ashangi  I  heard  that  the  anny  was  close  Xo 
Magdala^  and  might  return  at  any  moment ;  also  that 
the  fortress  was  much  farther  than  we  had  hitherto 
believed.  Although  I  had  but  little  faitli  in  any  imme- 
diate return,  still,  as  I  found  I  could  only  push  on  by 
leaving  all  imjyedimenta  behind,  I  left  at  Ashangi  my 
last  collector  and  my  tent,  and  went  on  with  only 
my  horses  and  one  mule  to  cany  provisions ;  for  the 
commissariat  stations  here  were  often  three  or  four 
marches  from  each  other,  and  no  food,  except  for  the 
animals,  coidd  be  obtained  at  the  intermediate  stages. 
I  then  started,  still  with  the  mule  convoy,  and  made 
a  double  march  to  Lat. 


78  PERSONAL  NARRATIFR 

The  rocod  led  round  the  Avestern  shore  of  the  lake  to 
the  southern  end,  and  then  traversed  a  broad,  fertile 
plain  of  black  soil,  in  which  stood  the  camping-ground 
of  Masagita,  On  the  lake-shore  Avere  an  unusual  number 
of  swallows  and  martins,  amongst  which  the  peculiar  black 
Psalidoprogne  pristoi)tera  was  conspicuous.  Leaving 
the  plain  of  Masagita,  our  road  ascended  a  wild,  narrow 
valley,  at  the  entrance  of  which  we  passed  a  fortified 
village,  surrounded  by  a  strong  wooden  palisade.  This 
border-land  between  Amliai^a  and  Galla,  Christian  and 
Mahommedan,  is  one  of  the  most  disturbed  parts  of 
Abyssinia,  and  more  robberies  and  murders  of  camp- 
followers  took  place  about  Ashangi  than  in  any  other 
portion  of  the  route.  From  the  valley  the  road  ascended 
to  a  stretch  of  flat  ground  at  about  10,000  feet,  the 
summit  of  the  Wombarat  pass.  Here  I  was  caught  in 
a  violent  hailstorm,  and  had  to  find  the  best  shelter  I 
could  behind  a  rock.  The  rain,  after  about  an  hour, 
abated,  and  I  rode  down  to  Lat,  the  path  being  so  slip- 
pery that  my  horse  came  down  with  me  more  than  once. 
At  Lat  I  put  up  with  a  friend.  Captain  Smith  of  the 
Commissaiiat,  but  the  rain  recommenced,  and  in  the 
middle  of  the  night  a  violent  gust  of  wind  blew  down 
our  abode,  a  light  single  bell  tent.  This  was  pleasant ; 
we  were  of  course  sleeping  on  the  ground,  and  were 
soon  wet.  Bell  tents  are  very  useful  in  some  respects ; 
they  keep  out  rain  very  fairly  as  long  as  they  stand ;  but 
in  a  high  wind  they  are  about  the  most  useless  invention 
conceivable.  The  little  pegs  used  have  no  hold,  and  pegs 
of  proper  length  are  too  lieaAy  to  be  carried,  as  the  tent 


LAT.  79 

i-equires  so  large  a  number.  The  Kttle  tents  used  on  the 
hills  in  India  are  lighter  than  bell  tents,  much  easier  to 
pitch,  and  afford  far  more  security  against  bad  weather. 

At  Lat  we  foimd  Major  Grant  and  Captain  Moore, 
who  had  been  sent  back  by  the  Commander-in-chief  to 
expedite  the  arrangements  for  the  carriage  of  commis- 
sariat stores  to  the  front.  The  Transport  Train  had,  from 
the  very  commencement  of  the  campaign,  proved  quite 
inadequate  by  itself  to  perform  this  service,  and  native 
caniage  had  been  very  largely  employed.  As  fai-  as 
Atala  all  had  worked  smoothly,  and  the  stores  had 
Ijeen  pretty  regularly  delivered  ;  but  between  Atala  and 
Ashangi,  and  still  more  beyond  Ashangi,  trickery  and 
oppression  on  the  part  of  some  of  the  petty  chieftains 
had  caused  a  complete  breakdown  in  the  system,  and 
had  not  Gobazye  s  chieftains  afforded  a  most  unexpected 
tmiount  of  assistance  to  the  officers  of  the  Commissariat 
in  procuring  supplies,  the  army  would  have  fared  far 
worse  than  it  did.  As  it  was,  there  was  always  abun- 
dance of  food,  though  it  only  consisted  of  tougli  beef 
and  inferior  flour. 

From  these  officers  we  at  last  learned  something 
definite  as  to  the  route  in  front,  and  the  great  passes  still 
to  be  surmounted.  They  had  quitted  the  head-quarters 
camp  at  Santara,  on  the  Wadela  plateau,  six  marches  of 
various  lengths  beyond  Lat,  and  Magdala  was  still  four 
or  five  marches  farther.  When  the  fortress  would  be 
reached  was  still  uncertain,  but  we  might  still  be  in  time 
to  witness  its  fall.  It  was  amusing,  with  reference  to 
this  subject,  to  hear  the  reports  which  circulated  fi'om 


80  PERSONAL  NARRATIVE. 

post  to  post.  The  mail  letters  were  carried  by  cavalry 
picquets,  and  the  men  were  always  eagerly  questioned 
for  news.  For  a  long  time  the  belief  in  aU  the  camps  in 
Tigr^  was,  that  Magdala  would  be  reached  and  assaulted 
on  the  25th  of  March,  and  some  of  the  Indian  sowars 
actually  told  us  that  it  had  been  captured  on  that  day, 
long  before  the  news  could  possibly  have  reached  them. 
Afterwards,  the  6th  of  April  was  the  appointed  date. 
As  is  well  known,  the  place  did  not  fall  tUl  the  1 3th. 

From  Lat  the  road  ascended  to  the  crest  of  the  Dafat 
pass,  which  is  not  so  high  as  that  of  Wombarat.  Hence 
again  there  is  a  fine  view  both  to  the  south,  over  the 
valleys  of  the  Tellari  river  and  its  feeders,  and  to  the 
eastward,  over  a  portion  of  the  same  great  valley  in  the 
Asulx)  Galla  country,  Avhich  was  seen  from  the  pass  noith 
of  Lake  AshangL  All  the  country  around  Lat,  as  far 
south  as  the  Abuyameda  range,  between  the  Tellari  and 
Takkazzy^  valleys,  consists  of  the  lower  group  of  traps. 
These  traps  are  more  or  less  inclined,  usually  at  low  angles, 
but  occasionally,  as  just  west  of  Lat,  Anth  high  dips. 
In  this  case  the  angle  is  45^  The  general  dip  is  to  the 
south. 

From  tlie  Dafat  pass  the  road  descends  into  a 
Avell- wooded  valley.  I  met  here  with  a  small  herd  of 
monkeys  {Cercopithecus  griseo-viindis),  which  are  not 
often  seen  at  quite  so  high  an  elevation.  I  also  met  with 
two  doves,  Peristera  afra  and  Turtur  semi-torqucUus, 
Euppell,  usually  met  with  at  a  lower  horizon. 

From  Marawa,  the  camping-ground  beyond  Lat,  to 
Dildi  on  the  Tellari,  here  a  very  small  brook,  was  a  long 


MARA  WA^DILDL  8 1 

march  of  sixteen  miles  over  a  series  of  steep  ascents  and 
descents,  forming  one  of  the  most  fatiguing  marches  of 
the  whole  route.  Of  the  convoy  of  mules  and  ponies 
with  which  I  marched  from  Senafd  to  Magdala,  more 
animals  fell  from  exhaustion  and  had  to  be  abandoned 
upon  this  march  than  on  any  other.  All  the  animals 
were  half  starved ;  grain  alone  had  been  procurable  at 
most  of  the  halting-places  we  had  recently  passed,  and 
the  poor  beasts  were  left  night  after  night  without  gi-ass 
or  hay.  Even  our  horses,  for  which  we  could  usually 
purchase  grass  from  the  Abyssinians,  were  getting  thin. 

The  character  of  the  valleys  about  Marawa  and  Dildi 
is  different  from  that  of  any  previously  traversed.  The 
country  is  a  network  of  wooded  glens,  without  any  large 
range.  All  the  sides  of  the  valleys  show  signs  of  agri- 
culture, all  have  evidently  been  cleared  in  patches,  and 
cultivated  for  a  season  or  two,  and  then  abandoned,  a 
common  system  in  wild  countries,  where  the  land  is 
largely  in  excess  of  the  wants  of  the  population.  Bushes 
and  trees  spring  up  again  in  the  abandoned  fields,  but 
grow  in  plots,  which  still  show  the  limits  of  the  clearing. 
A  peculiar  appearance  is  thus  given  to  the  landscape ; 
from  a  distance  it  almost  looks  as  if  the  sides  of  the 
valleys  were  covered  by  fields  with  hedges,  as  in 
Ungland. 

At  Dildi  there  was  a  comparatively  large  camp,  much 
baggage  having  been  left  here.  I  started  the  next  morn- 
ing with  Captain  Arbuthnot,  an  aide-de-camp  of  Sir  R. 
Napier,  intending  to  make  a  double  march  and  reach  the 
Takkazzy^.    The  road  from  Dildi  ascended  a  steep  incline 

G 


82  PERSONAL  NARRATIVE. 

to  the  crest  of  tlie  Wandaj  pass,  10,500  feet  above  the 
sea.  This  pass  crosses  the  Abuyameda  range,  the  flat- 
topped  masses  of  which,  right  and  left  of  the  pass, 
rise  to  12,000  feet  or  more.  All  the  upper  portion 
of  this  range  consists  of  the  higher  group  of  traps,  hori- 
zontal as  usual,  and  with  the  strongly-marked  vertical 
scarps  eharacteiistic  of  the  trachyte  beds,  which,  how- 
ever, at  the  pass  itself,  are  less  numerous  than  else- 
where. A  peculiar  feature  near  the  crest  of  the  pass  is 
the  presence  of  very  many  large  dykes  of  basalt,  which 
weather  more  slowly  than  the  rocks  around,  and  stand 
out  in  relief  like  gigantic  walls  ui>on  the  hill-side.  Their 
direction  varies  considerably ;  some  of  the  most  prominent 
run  nearly  north-west  and  south-east. 

The  vegetation  at  the  crest  of  this  pass  is  rendered 
striking  by  the  abundance  of  the  remarkable  Tupa 
rhyiicophetalumy  a  great  spike  of  leaves,  not  unlike 
those  of  tobacco,  on  the  top  of  a  high  stalk.  I  here  first 
saw  one  or  two  birds,  such  as  Pratincolu  sordida,  and 
the  black  and  yellow  weaver  bird,  EupUcte^  xantJwmelaSy 
peculiar  to  high  altitudes.  Nectarinia  Tcdcazze  Avas 
common,  and  the  large  francolin,  or  spur-fowl,  as  it  was 
called  in  the  army,  F.  Erkelii^  abounded.  The  ground 
at  the  summit  of  the  pass  Avas  burrowed  in  every  direc- 
tion by  a  peculiar  kind  of  field  rat,  Mus  ahyssinicus 
of  Eiippell. 

Before  reaching  the  summit  we  were  overtaken  by 
heavy  mist  and  rain,  and  it  poured  nearly  the  whole 
afternoon.  It  was  dangerous  to  go  on ;  the  path  down 
was  too  slippeiy  for  the  animals  to  keep  their  feet ;  they 


WANDAJ  PASS— HORSE  STOLEN.  8  3 

had  already  slipped  down  more  than  once,  and  a  horse  of 

mine  had  nearly  fallen  over  the  steep  hill-side  on  the 

way  up,  while  the  descent  to  the  south  was  represented 

to  us  as  much  steeper,  and  with  a  precipice  beside  it. 

The  carcases  of  mules  and  horses  showed  how  many  had 

fallen  on  this  ascent.     I  had  no  tent ;   Captain  Smith 

and  I  had  arranged  to  march  to  the  front  together,  and 

to  use  his  tent,  which  was  the  lightest,  but  he  had  been 

detained  at  Lat   However,  I  found  shelter  with  a  friend. 

The  heavy  rain  continued  nearly  all  night,  a  high  wind 

blew,  and  it  was  bitterly  cold,  and  I  came  out  in  the 

morning  expecting  to  find  some,  both  of  the  poor  Indian 

camp-followers,  who  had  been  out  in  the  rain  all  night,  and 

of  the  animals,  dead  with  cold  and  wet.   To  my  surprise, 

Hone  seemed  to  have  suflTered.     As  no  grass  had  been 

procurable,  the  animals  were  all  turned  out  to  feed  on 

the  fine  wiry  herbage  of  the  mountain,  and  I  sent  my 

liorses  with  them.     I  had  but  one  man  to  look  after  two 

horses  and  a  mule,  and  whilst  he  was  bringing  back  the 

latter,  which  had  strayed  slightly,  a  chestnut  Ai\ab,  which 

I  had  hitherto  ridden,  and  a  very  docile  useful  animal, 

<-lisappeared.     After  a  short  time  I  discovered  its  tracks, 

and,  only  waiting  to  get  a  gun,  followed  them.     The 

tracks  of  a  man  leading  the  horse  Avere  plain  :  he  had 

.stolen  the  beast,  and  led  him  round  the  hill,  keeping 

just  out  of  sight  of  camp,  and  had  then  gone  to  the 

westward.    After  a  mile  or  two  I  lost  the  tracks  amongst 

a  number  of  others,  and  had  to  give  up  the  chase.     Of 

f'ourse  I  never  heard  anything  more  of  my  horse.     Such 

thefts  were  common,  and  several  oflBcers  lost  horses.    To 

g2 


S4  PERSOXAL  XJHRJTirK 

complete  my  misfortune,  whilst  I  was  searching  for  the 
horse  my  saddle  was  also  stolen  from  camp. 

After  delaying  the  search  until  I  had  only  just  time  to 
get  down  to  the  Takkazzy^  by  nightfall,  I  started,  and 
walked  down  the  steep  incline,  leading  my  remaining 
horse,  who  carried  my  bed,  &c.  It  had  begun  to  rain 
again,  but  half-way  down  the  weather  cleared  up,  the 
rain-storms  evidently  keeping  much  to  the  higher  ranges. 
There  was  a  camping-place  half-way  down  at  Muja,  but 
I  went  on  to  the  "bottom,  3,000  feet  below  Wandaj, 
crossed  the  Takkazzy^,  here  a  small  stream,  and  caught 
up  the  party  with  Roddy's  mule-train  once  more  at  the 
camping-ground. 

On  the  next  day  we  ascended  the  steep  road  leading 
to  Santara  on  the  Wadela  plateau,  and  thence  went  on 
to  Gaso,  another  camp  twelve  miles  further.  There  is 
no  very  distinct  stratification  in  the  Takkazzyd  valley  ; 
and  as  I  did  not  clearly  make  out  the  distinction  be- 
tween the  two  groups  of  traps  till  after  passing  it  on  my 
return  from  Magdala,  I  omitted  to  notice  if  the  lower 
series  occurs  there.  Almost  without  doubt  it  must  do 
so,  but  the  ix)ad  reascends  to  the  horizontal  trachytic 
series  on  the  great  scarp  south  of  the  Takkazzyd. 

The  march  from  Santara  to  Gaso,  and  thence  to 
Yasendyd  and  Bethor,  led  across  the  Wadela  plateau,^  at 
first  at  an  elevation  of  at  least  10,500  feet ;  subsequently, 
south  of  Gaso,  at  a  somewhat  lower  level,  Yasendye  and 
Bethor  being  about  9,500  feet  above  the  sea.     In  the 

^  In  fact,  all  the  northern  portion  belongs  to  the  district  of  Dalanta,  and 
only  the  south-weatem  portion,  from  Gaso  to  Bethor,  to  Wadela. 


TAVNA  OF  JTADELA  PLATEAU.  85 

nortliem  portion,  from  Santara  till  about  six  miles  south 
of  Gaso,  all  the  rocks  exposed  consist  of  trachyte  and 
trachytic  ash,  containing  large  blocks  of  black  jet-like 
obsidian,  which  some  of  the  soldiers  took  for  coal  They 
soon  found  out  their  mistake  when  they  tried  to  burn  it. 
The  scenery  is  very  peculiar.  Hummocks  of  grey  rocks 
covered  with  lichens  stand  up  from  amidst  the  fine  wiry 
grass  already  mentioned  as  occurring  on  the  high  passes 
of  Ferra  and  Wandaj,  and  alternate  with  small  marshy 
flats,  the  soil  in  which  is  an  imperfect  peat  The  fauna 
is  mostly  different  from  that  of  lower  elevations.  The 
handsome  white-breasted  buzzard  {Buteo  augur),  the 
dwarf  goose  {Bemicla  cyanoptera),  and  the  caruncled 
ibis  {Harpiprion  canwculata),  are  very  abundant.  A 
duck  found  also  at  lower  levels  {AncLS  Jlavirostris)  occurs 
in  the  streams  and  marshes.  Two  peculiar  chats  (Saxicola 
frenata,  Heugl.  and  Pratiiicola  sordida),  the  yellow- 
throated  pipit  {Macronyx  JlavicolUs),  a  species  of  finch 
allied  to  the  canary  bird  {CHthagra  iiigriceps),  a  wat- 
tled lapwing  (Lobivanellus  melanocephalus),  and,  above 
all,  the  red-legged  chough  {Pyrrhocorax  alpinus)^  are 
peculiar  and  characteristic  forms.  The  Abyssinian  lam- 
mergeyer  and  the  thick-billed  crow  {Corvultur  crassi- 
rostris)  were  also  frequently  seen.  Mammals  were 
scarce ;  the  only  conspicuous  one  was  the  little  Mas 
abyssinicuSy  which  was  most  abundant  on  the  higher 
parts  of  the  plateau. 

Several  of  these  birds  disappear  below  10,000  feet  on 
the  portion  of  the  plateau  near  Yasendy^  and  Bethor, 
where  the  rocks  exposed  are  chiefly  basaltic,  and,  as  is 


86  PERSONAL  NARRATIVE. 

usual  under  such  circumstances,  a  black  soil  prevails, 
evidently  highly  favourable  to  the  growtli  of  cereals,  and 
widely  cultivated  The  wild  grasses  are  quite  different 
from  those  on  the  trachyte. 

At  Yasendyd  I  was  surprised  at  seeing  the  white  and 
black  crow  {Corviis  scapulatus),  which  I  had  never 
before  met  with  since  leaving  the  flat  coast  region  near 
Komayli.  Here  it  associated  with  Corvultur  and  Comes 
affinis. 

I  had  walked  the  long  march  from  Gaso  to  Yasendy6, 
the  greater  portion  through  pouring  rain.  On  my  arrival, 
however,  I  was  most  hospitably  received  by  my  old 
friend  Major  Bardin.  The  next  morning  I  determined 
to  ride  my  remaining  horse  with  an  Otago  saddle, 
taking  only  a  rug  with  me  by  way  of  bed,  but  on 
mounting  the  animal  I  found  he  was  lame.  Major 
Bardin  very  kindly  lent  me  a  pony,  but  I  did  not  make 
much  progress. 

The  road  from  Yasendy^  to  Bethor  is  precisely  similar 
to  that  north  of  the  former  place,  over  an  undulating 
plateau,  with  much  cultivation  and  long  grass.  In  one 
or  two  places  before  reaching  Yasendy^,  deep  ravines, 
exposing  cliffs  of  basalt,  had  been  seen  to  the  left  of  the 
road,  and  a  couple  of  miles  beyond  Bethor  we  came 
suddenly  on  the  brink  of  the  mighty  chasm  in  which 
the  Jitta  river  runs,  and  to  which  all  the  minor  valleys 
converge. 

Of  all  the  grand  scenery  met  with  in  Abyssinia,  none 
equalled  this  wonderful  gorge.  It  is  3,500  feet  deep, 
and  looks  scarcely  a  mile  across.    The  sides  are  extremely 


JITTA  RAVINE.  87 

steep,  in  places  nearly  perpendicular.  The  horizontal 
beds  on  both  sides  appear  to  correspond  exactly ;  half- 
way down  there  is  a  well-marked  terrace,  evidently 
formed  by  the  same  bed,  on  both  sides  of  the  river.  The 
bedding  is  very  distinct. 

It  is  to  be  regretted  that  all  those  geologists  who  dis- 
believe in  the  power  of  running  water,  and  appeal  to 
such  Dei  ex  machind  as  marine  action  and  waves  of 
translation,  cannot  see  a  few  such  marks  of  the  handi- 
work of  rain  and  rivers  as  are  shown  in  these  gorges. 
There  is  not  the  slightest  possibility,  so  far  as  I  can  see, 
of  explaining  their  origin  by  any  other  force  than  that 
of  the  streams  flowing  in  them.  Any  faulting  or  dis- 
location of  the  rocks  is  out  of  the  question,  there  is  no 
evidence  of  glacier  action  at  any  past  time,  and  even  the 
most  fervent  apostle  of  marine  denudation  would  scarcely 
credit  it  with  the  formation  of  a  Titanic  trench  three- 
quarters  of  a  mile  deep  aud  very  little  more  in  breadth. 

At  the  bottom  of  the  ravine  ran  a  beautifully  clear 
stream  in  a  pebbly  bed.  Some  of  our  party  pushed  on 
to  the  Dalanta  plateau ;  I  remained  for  the  night  with 
Captain  Roddy  in  th^  valley.  The  temperature  at  the 
bottom,  only  6,000  feet  above  the  sea,  was  a  most 
agreeable  change  from  that  of  the  bleak  plateau.  On 
the  following  morning,  having  the  loan  not  only  of  a 
pony  but  of  a  saddle  also,  I  considered  myself  fairly 
equipped  once  more,  and  started  for  Dalanta.  The 
ascent,  like  the  descent  of  the  previous  day,  was  by  the 
road  which  King  Theodore  had  constructed  for  the  trans- 
port of  his  artillery  to  Magdala,  a  broad  path,  at  least 


88  PERSONAL  NARRATIFE, 

twenty  feet  wide  in  most  places,  and  thoroughly  cleared, 
but  excessively  steep  in  parts — a  wonderful  record  of  the 
man's  perseverance  and  intelligence,  though  not  of  his 
engineering  skill  The  section  exposed  by  the  road 
gave  an  excellent  idea  of  the  rocks.  The  majority  are 
basaltic ;  a  little  columnar  trachyte  occurs  near  the 
summit  north  of  the  ravine,  and  on  the  south  side,  just 
below  the  top  of  the  scarp,  the  following  beds  occur  in 
descending  order : — 

1.  Compact  basalt. 

2.  White    argillaceous   rock    of  sedimentary   origin, 

twenty  to  twenty-five  feet  thick. 

3.  Volcanic  ash  and  basalt. 

4.  Coarse  brown  sandstone,  fifty  to  sixty  feet, 

5.  Trachyte  with  crystals  of  ghissy  felspar. 

6.  Basalt. 

The  sedimentary  beds  can  be  traced  by  the  eye  for  a 
short  distance  along  the  scarp,  the  sandstone  being  very 
conspicuous,  but  they  have  evidently  no  great  extension, 
as  they  do  not  recur  on  the  north  side  of  the  mvine. 
Below  the  above,  about  half-way  down  to  the  river,  some 
black  shales  are  interstratified  in  the  traps.  I  could 
find  no  organic  remains,  but  fossil  wood  is  said  to  have 
been  met  with  here. 

The  road  from  the  Jitta  to  the  Bashilo  crosses  a  portion 
of  the  Dalanta  plateau,  which  is  even  flatter  than  the 
Wadela,  and  consists,  like  the  southern  portion  of  the 
latter,  of  basaltic  rock  covered  with  black  soil,  and 
mostly  cultivated.     After  traversing  this  for  about  six 


ARRIFJl  BEFORB  UJODJLA.  89 

miles,  we  reached  the  scarp  of  the  Baahilo  raviDc,  even 
deeper  than  that  of  the  Jitta,  but  less  striking,  because 
Dot  so  abrupt ;  the  opposite  side,  insteiid  of  being  one 
immeDde  unbroken  cliif  like  that  on  both  sides  of  the 
Jitta,  being  cut  up  by  ravines  and  of  unequal  height. 
Opposite  to  us  rose  the  fortress  of  Magdala,'  the  scarp 
Burrounding  which  looked  small  at  this  distance,  and 
dwarfed  by  the  higher  fiat-topped  hills  beyond.  On  a 
ridge  to  the  right,  the  two  British  camps  were  con- 
spicuous, that  of  the  Ist  Brigade  on  the  Arog}'^  spur, 
and  that  of  the  2d  Brigade  a  little  nearer  and  lower. 
These  heights  had  been  occupied  the  evening  before,  and 
the  smart  skirmish  which  ensued  with  Theodore's  army 
had  l>een  distinctly  seen  from  Dalanta ;  but  although 
the  flashes  and  smoke  of  the  small  arms  and  steel 
guns  had  been  clearly  diatiuguisbable,  the  only  sounds 
heard  were  those  made  by  Theodore's  large  cannon,  fired 
from  Fala, 

I  descended  the  steep  road  to  the  river,  over  rocks 
similar  to  those  seen  in  the  Jitta  sections,  but  without 
sedimentary  iuterstratifications ;  and,  reascending  the 
opposite  hills,  reached  the  camp  at  Arogyi5  before  even- 
ing. This  was  on  Saturday,  the  10th  of  April,  the 
evening  on  which  the  prisoners  were  released.  I  had 
marched  from  Senaf^,  with  only  one  day's  halt,  in  twenty- 
three  days,  and,  of  course,  anything  like  a  careful  exa- 
mination of  the  country  was  out  of  the  question, 

■  It  uuwell  to  mention  that  the  accent  in  this  word  is  on  the  fintsjiltible. 
Magdila  ia  inconect  This  is  mentioned  bj  MaTldiam  also,  but  the  false 
■eaentnation  baa  been  very  genenUy  adopted,  i 


90  PERSONAL  NJRRATIFR 

It  is  quite  foreign  to  the  piiri)ose  of  the  present  work 
to  describe  the  capture  of  MagdaJa,  which  took  place 
two  days  after  my  arrival.  The  remarkable  good  fortune 
which  crowned  the  expedition  was  far  less  favourable  to 
scientific  inquiry  than  to  military  renown ;  for  the  sub- 
sequent rapid  retreat  to  the  coast,  however  brilliant  as 
a  military  manoeuvre,  entirely  prevented  me  from  exa- 
mining the  country  in  any  detail. 

All  the  hills  around  Magdala  appear  to  be  of  horizontal 
traps,  chiefly  basalt,  although  trachyte  also  occurs,  and 
the  plateau  of  the  fortress  itself  consists  of  the  latter 
rock ;  the  steep  surrounding  scarp,  which  would  have 
rendered  the  place  so  diflScult  of  capture  had  it  been 
defended,  being  the  ordinary  form  of  outcrop,  already 
alluded  to  as  characteristic  of  the  trachyte  beds.  Inter- 
stratifications  of  white  shale,  much  hardened  as  if  by 
lava-flows,  were  met  with  in  two  or  three  places  around 
Arogy^,  and  were  especially  noticed  close  to  the  camp  of 
the  2d  Brigade ;  but  no  organic  remains  could  be  de- 
tected in  them. 

The  camp  before  Magdala  was  certainly  the  most  un- 
pleasant abode  in  which  I  found  myself  in  Abyssinia. 
The  whole  force  was  crowded  into  the  smallest  possible 
space,  in  small  thin  tents,  not  nearly  large  enough  even 
to  hold  the  men;  the  only  food  was  inferior  flour  and 
tough  beef,  and  it  was  difticult  to  obtain  suflScient  water 
to  drink ;  washing  was  a  luxury  only  to  be  occasionally 
attained.  It  was  decidedly  worse  than  at  Malkatto. 
Water  there  was  scarce,  but  a  swim  in  the  sea  was 
always  a  resource  available  ;  and  there  was  no  crowding  ; 


ATTEMPTS  AT  FURTHER  EXPLORATION.  9 1 

while  as  regards  the  Commissariat,  I  can  only  say  that  I 
wish  it  had  a  dep6t  in  many  of  the  wilder  parts  of  India, 
and  that  I  were  entitled  to  draw  rations  as  in  Abyssinia. 
I  have  often  fared  far  worse  in  India,  with  all  the  re- 
sources of  the  country  at  my  command.  It  is  not  sur- 
prising that  all  the  troops  at  Magdala,  who  had  for  two 
months  been  working  hard,  and  for  at  least  a  month 
been  faring  hard  also,  were  delighted  at  the  prospect  of 
leaving  the  country. 

I,  on  the  other  hand,  was  disgusted  at  being  unable 
to  advance  farther  into  Abyssinia.  Above  all,  I  had 
hoped  to  have  thoroughly  explored  the  great  lake  of 
Dembea  or  Tsana,  which  has  never  been  accurately 
mapped.  I  believe  that  there  are  few  geological  prob- 
lems more  interesting  than  those  connected  with  the 
origin  of  lake  basins,  and  those  of  Africa,  tlic  only  great 
lakes  in  the  tropics,  are  of  peculiar  importance.  But  Sir 
Robert  Napier  would  not  listen  to  any  proposals  of  ex- 
ploration. He  even  refused  permission  to  Mr,  Munzinger 
to  go  into  the  country  of  the  Asubo  Gallas,  although 
one  of  the  principal  chiefs  had  guaranteed  his  safety, 
and  although  Mr.  Muuzinger's  loDg  experience  and  great 
knowledge  of  the  Abyssinian  people  would  have  alone 
been  almost  suflScient  to  secure  him  from  injury.  A 
few  days  after  leaving  Magdala,  in  order  to  leave  do 
means  untried,  I  applied  to  be  allowed,  with  a  small 
party,  to  go  to  Lake  Tsana,  and  thence  march  to  the 
south-west  towards  Gondokoro,  endeavouring  to  explore 
the  Sobat  river,  but  without  any  success.  Sir  Eobert 
was  determined  to  have  every  European  over  whom  he 


92  PERSONAL  NARRATIVE, 

could  in  any  way  exercise  authority  out  of  the  country 
in  as  short  a  time  as  possible.^ 

I  left  the  camp  before  Magdala  on  the  1 6th  of  April, 
three  days  after  the  fall  of  the  fortress,  and  marched 
back  as  rapidly  as  I  could  to  the  Wadela  plateau.  Here 
I  halted  for  a  day  or  two  to  obtain  a  few  specimens  of 
the  birds,  of  which,  however,  I  could  not  collect  many, 
as  I  had  to  prepai'c  the  skins  myself  At  Yasendye, 
JVIr.  Markham,  the  Honoraiy  Secretary  of  the  Geogra- 
phical Society,  who  had  been  with  the  expedition 
fix)m  the  beginning,  but  whom  I  had  never  met  tiU 
I  reached  Magdala,  came  up  from  the  front  with  Major 
Grant.  He  had  no  hope  of  penetrating  any  further  into 
the  country,  and  was  on  his  road  back  to  England.  He 
left  with  me  an  aneroid  and  a  boiling-point  thermometer, 
in  case  I  should  be  more  successful. 

From  Yasendyd  I  came  on  to  Gaso  and  Santara.  I 
had  some  days  previously  written  to  Ashangi  for  my 
tent  and  collecting  apparatus  to  be  sent  on,  as  I  wished  to 
procure  specimens  of  the  birds  and  mammals  peculiar  to 
this  elevated  region.  I  found,  however,  that  my  tent, 
&c.  had  not  arrived,  and  that  the  whole  camp  had  been 
removed  from  Santara  to  the  Takkazzy^  except  the  tent 
of  the  guai'd.  There  was  no  place  in  which  to  remain, 
for  the  guard-tent  was  of  course  crowded,  and  very 
reluctantly  I  was  compelled  to  leave  this  most  promising 

>  I  do  not  for  an  instant  doubt  that  Lord  Napier  was  not  only  perfectly 
justified  in  refusing  permission  to  any  one  to  leave  the  line  of  march,  but  I  fear 
so  little  interest  is  taken  in  England  in  any  exploration  for  other  than  purely 
commercial  motives,  that  he  would  have  been  blamed  had  he  permitted  any. 


RETURN  TO  LAKE  aSHANGI.  93 

ground  with  only  about  seven  species  of  animals  not 
obtained  elsewhere.  These  comprised  a  peculiar  hyrax, 
the  little  Mus  ahyssinicus,  Saxlcol(tfrenata,  Pratincola 
sordiduy  the  rare  Lobivanellus  melanocephaluSy  Macronyx 
jlavicollisy  and  Euplectes  xanthonielas.  Had  I  been 
al)le  to  devote  more  time  to  the  exploration  of  these  hills, 
I  might  doubtless  have  increased  the  number  consider- 
ably, and  have  obtained  a  larger  proportion  still  of  the 
peculiar  forms  described  by  Rtippell  and  Von  Heuglin 
from  the  mountains  of  Samyen. 

This  being  out  of  the  question,  however,  I  pushed  on  to 
Lake  Ashangi.  On  my  road  I  obtained  two  or  three  more 
8j>ecimens  of  birds  on  the  top  of  Wandaj,  including 
Falco  tanypterus,  which  I  had  frequently  seen  l^efore  but 
never  secured.  I  also  saw  Hypotinorchis  concolor,  the 
slate-coloured  merlin,  of  which  I  did  not  secure  a  speci- 
men. Close  to  Dildi  I  shot  one  of  the  mouse-coloured 
antelopes,  Cephalophus  madoqua. 

The  released  prisoners  from  Magdala  were  on  their 
march  towards  the  coast,  and  I  met  several  of  them, 
amongst  them  Dr.  Schimper,  the  veteran  naturalist,  who 
has  passed  forty  years  in  exploring  the  botany,  zoology, 
and  geology  of  the  country.  He  was  a  fellow-student 
with  Agassiz,  and  having  remained  so  long  away  from 
European  changes  of  thought,  has  retained  the  geological 
ideas  of  a  past  generation,  differing  from  those  of  the 
present  less  perhaps  in  reality  than  in  nomenclature,  so 
that  I  found  some  difficulty  in  understanding  his  views. 
Although,  so  far  as  I  am  aware,  he  has  published  nothmg 
himself,  he  has  supplied  many  successive  travellers  with 


94  PERSON  J  L  NARRATIVE. 

infonnation,  and  has  indirectly  contributed  more  perhaps 
than  any  other  living  European  in  making  the  fauna 
and  flora  of  the  Ethiopian  highlands  known  to  the  scien- 
tific world  of  Europe. 

I  reached  Lake  Ashangi^  once  more  on  April  29th, 
and  remained  there  for  seven  days,  collecting.  Lieut. 
St.  John  had  lent  me  an  india-rubber  boat,  which  was 
most  useful,  as  no  native  craft  of  any  kind  existed 
on  the  water.  The  amount  of  astonishment  which  the 
lx)at  produced  amongst  the  Abyssinians  was  naturally 
great  The  lake  swarmed  with  water-birds,  but  many 
kinds  of  ducks  had  left.  I  saw  the  shoveller,  always 
one  of  the  latest  to  migrate  to  the  north,  and  shot 
the  created  pochard  {Fuliguh  cristata),  but  there 
appeared  to  be  decidedly  fewer  than  when  I  passed  on 
my  way  to  Magdala  a  month  earlier.  The  Eg}"ptian 
goose  {Chenalopex  (eyyptiaca)  occurred  in  large  num- 
bers. The  most  abundant  bird  on  the  lake  was  the 
crested  grebe  {Podiceps  (^ristata) ;  I  also  shot  the  eared 
grebe  (P.  aurita)^  which  was  scarce,  and  the  common 
dabchick  (P.  minor),  A  small  cormorant  {Graculus 
africanus)  was  occasionally  met  with,  and  I  once  saw  a 
white  pelican,  which,  however,  I  failed  to  secure.  The 
crested  coot  {FuUca  cristata)  was  tolerably  abundant 
The  umbre  {Scopus  xnnhretta)  occurred  on  the  margin  of 
the  lake,  with  a  large  black  and  white  ibis,  of  which  I 
failed  to  obtain  a  specimen.  In  the  mai-shes  on  the 
banks  Rallus  Rougeti  and  Gallinula  chloropii^  occurred, 

^  A  brief  description  of  this  lake  will  be  found  in  a  subsequent  page  in  the 
part  devoted  to  the  Geology,  Section  I. 


FAUNA  OF  ASHANGL  95 

and  snipe  abounded.  Flocks  of  Ibis  carunculata  con- 
sorted with  the  Egyptian  geese  on  the  meadows  around 
the  lake,  but  both  had  become  very  wild,  owing  to 
their  being  so  frequently  fired  at.  The  spur-winged 
plover  {Hoplopterus  spinosus)  was  met  with  occasionally 
in  the  same  places  or  along  the  shore. 

Amongst  the  high  grass  which  grew  in  places  near 
the  lake  I  found,  rather  to  my  surprise,  large  flocks 
of  the  black  and  yellow  weaver-bird  {Euplectes  xantho- 
melas),  which  elsewhere  I  had  only  seen  at  much  greater 
elevations.  On  the  stunted  trees,  the  beautiful  long-tailed 
roller  {Coractas  abysstnica)  was  far  from  rare,  and  the 
iridescent  starling  {LamprocoUns  chalyhceus)  was  fre- 
quently met  with.  Amongst  the  other  birds  here  col- 
lected were  Pratincola  semitwquatcty  Heug.,  Turdus 
oUvacinics,  Bp.,  Bradyornis  clwcolatina,  Cotyle  cincta, 
and  C.  minor.  I  also  obtained  a  single  specimen  of  the 
secretary  bird. 

The  meadow  in  which  the  camp  stood  was  dotted  over 
by  heaps  of  earth,  exactly  like  mole-hills,  thrown  up  by 
Bathyergus  splendenSy  a  burrowing  short-tailed  rodent, 
the  size  of  a  large  Norway  rat,  but  resembling  in  appear- 
ance the  bamboo  rats  (Rhizomys)  of  the  Himalayas  and 
BuiTOese  countries.  I  kept  one  alive  for  two  or  three 
days,  feeding  it  on  roots  of  grass,  but  it  finally  escaped. 
There  were  unusually  few  jackals  and  hyaenas.  A  few 
koodoo  were  said  to  occur  in  the  neighbouring  hills,  and 
I  found  the  tracks  of  pigs  and  otters  along  the  shore  of  the 
lake.  There  were  a  few  land-shells  to  be  found,  Bulimus 
Olivleriy  a  species  of  Vitrina,  and  an  JEnnea,  the  two 


1)6  PEHHONAL  NARRATIVE, 

latter  undescribed,  being  the  most  common.  Despite  the 
large  number  of  fish-eating  birds  on  the  lake,  all  the  Abys- 
sinians  persisted  in  denying  the  existence  of  fish  in  the 
waters.  I  only  saw  one  small  kind,  but  the  occurrence 
of  otters  alone  serves  to  indicate  the  existence  of  some 
larger  forms.     I  could  catch  none,  however. 

The  head-quarters,  now  the  rear-guard  of  the  army, 
reached  Ashangi  on  the  5th  and  left  on  the  6th  May, 
abandoning  such  stores  as  could  not  be  carried  with 
them,  and  making  a  bonfire  of  huts,  empty  packages, 
all  the  hay  not  wanted,  &c.  ;  400  or  500  mules  and 
ponies  which  were  either  lame  or  seriously  galled,  so 
that  there  was  no  probability  of  their  being  made  of 
service  during  the  return  journey  as  pack  animals,  were  ' 
also  abandoned,  and  a  great  scramble  took  place  amongst 
the  Abyssinians  for  their  possession.  The  scene  was  most 
amusing.  Crowds  of  the  poorer  Abyssinians  hung 
about  the  camp,  picking  up  anything  they  could,  and  by 
no  means  disdaining  to  pilfer  if  a  chance  afforded.  Now 
and  then  a  few  would  be  driven  off,  and  the  whole  would 
take  to  flight  for  the  moment ;  soon  returning,  however. 
As  each  corps  in  succession  moved  off  its  ground  with 
its  baggage,  these  people  crowded  into  the  abandoned 
space  to  search  for  anything  which  might  have  been  left 
behind,  and  they  might  be  seen  carrying  off  the  most 
singular  articles, — empty  packages,  broken  Otago  saddles, 
or  fragments  of  canvas,  blocks  of  firewood,  and  bundles 
of  grass.  A  similar  scene  took  place  at  every  camp, 
varied  occasionally  by  some  valuable  commissariat  stores, 
such  as  flour  or  grain,  being  left  behind  for  want  of 


RETURN  MARCH  TO  ANTALO,  9  7 

carriage,  and  by  a  free    fight   taking   place   for  their 
possession. 

The  march  back  demands  but  brief  notice.  Finding  it 
most  tedious  to  march  on  the  narrow  steep  roads  with 
the  head-quarters  camp,  owing  to  the  lengthy  train  of 
baggage — which  started  generally  at  daybreak,  and  the 
last  of  which  rarely  accomplished  a  march  of  eight  or 
ten  miles  before  dark — Dr.  Cook  and  I  made  a  double 
march  from  Makhan  to  Atala,  and  afterwards  kept  with  a 
small  force  which  marched  one  day  ahead  of  the  Com- 
mander-in-chiefs camp.  We  reached  Antalo  once  more 
on  the  11th,  and  I  halted  there  a  day.  St.  John,  who 
had  brought  the  telegraph  thus  far,  had  explored  the 
neighbourhood,  and  carefully  marked  down  a  couple  of 
owls  {Bubo  ci7ierascens),  which  inhabited  some  rocks  by 
the  Buya  stream,  about  two  miles  from  camp.  Besides 
killing  one  of  these,  I  obtained  several  other  rare 
birds  on  the  same  day,  amongst  them  a  pair  of  Ibis 
coniata. 

As  I  heard  that  the  head-quarters  and  rear-guard 
would  only  halt  for  two  days  at  Buya,  I  left  on  the 
13th  May,  in  order  to  have,  in  any  case,  a  couple  of  days 
for  the  examination  of  the  junction  of  the  limestone  and 
sandstone,  and  to  search  for  fossils  at  Agula  and  Dongolo. 
I  could  not  even  spare  a  day  to  visit  Antalo  town  and 
Chelikot.  Nothing  worthy  of  record  occurred  on  the 
road  to  Agula.  The  weather  had  become  fine  again,  and 
the  road  being  comparatively  good  and  hard  there  w^as 
but  little  delay  in  marching.  Moreover,  all  commissariat 
supplies,  and  especially  grass  and  grain  for  the  animals, 

H 


98  PERSONAL  NARRATIVE, 

being  now  abundant,  all  the  mules  and  horses  were 
rapidly  getting  into  good  condition,  and  much  fewer 
fell  or  knocked  up  on  the  road.  Still  I  frequently  pre- 
ferred allowing  the  whole  camp  to  go  ahead,  and  march- 
ing quietly  by  myself  in  the  afternoon,  so  that  I  could 
examine  the  rocks  along  the  road,  or  shoot  any  rare  birds 
or  animals  I  might  see.  This  was,  I  believe,  against 
orders,  as  there  was  supposed  to  be  risk  of  being 
attacked,  but  the  risk  was  very  small ;  all  that  was 
necessary  was  to  keep  the  mules  in  sight,  as  otherwise 
they  might  have  been  plundered  by  some  of  the  petty 
thieves  who  hung  about  the  route  of  the  army.  Much 
more  was  related  about  thieves  than  was  true.  Petty 
pilfering  took  place  everywhere,  as  amongst  all  civilized 
and  semi-civilized  peoples ;  but,  except  in  the  disturbed 
country  around  Ashangi,  most  of  the  stories  of  attacks 
and  thefts  were  invented  by  muleteers  and  camp  fol- 
lowers to  conceal  their  own  vUlanies.^  I  can  only  say, 
in  justice  to  the  Abyssinians,  that  both  in  marching  to 
Magdala  and  in  retm-ning  I  rode  or  walked  the  gi'eater 
portion  of  the  distance  either  entirely  alone  or  Avith  one 
or  two  companions,  who,  like  myself,  preferred  keeping 

^  An  admirable  instance  occurred  at  Undul  Wells,  when  I  was  staying  there. 
Some  muleteers  arrived  with  three  or  four  Shoho  prisoners  bound,  and  asserted 
that  these  men  had  attacked  them,  and  had  attempted  to  rob  their  mules  on 
the  road.  The  Shohos,  on  the  other  hand,  declared  that  they  had  been  con- 
veying stores  from  Zulla  to  Senuft^  for  the  Commissariat,  and  were  returning 
with  the  money  when  attacked  by  the  muleteers.  On  inquiry  this  turned 
out  to  be  the  case,  the  money  taken  from  the  Shohos  was  found  upon  the 
muleteers.  Another  trick  of  the  muleteers  was  to  sell  stores  entrusted  to 
them  to  Shohos  or  Abyssinians,  and  to  account  for  the  deficiency  by  stories 
of  robbeiy.  Undoubtedly  a  few  cases  of  robbery  did  take  place,  but  the 
greater  number  of  asserted  instances  were,  I  believe,  imaginary. 


OOLITIC  FOSSILS  NEAR  AG  ULA.  9  9 

out  of  a  crowd,  and  that  I  never  met  with  anything 
but  friendly  greetings  from  the  people  who  passed.  They 
were  especially  amicable  in  Tigrd.  The  usual  conversa- 
tion was  something  of  the  following  kind  : — 

Ahijssinian  (producing  a  small  piece  of  cord,  originally 
blue,  now  of  various  colours,  from  around  his  neck)  : 
"  Christi  ?"  which  being  inteipreted  was  supposed  to  be 
an  inquiry  as  to  the  Englishman's  Christianity. 

Englishrtian :  "  Yes,"  with  a  nod. 

Abyssinian :  **  Taib "  (Arabic  for  *'  good,"  probably 
the  whole  of  the  speaker's  attainment  in  the  language). 

The  blue  cord  is  in  Abyssinia  the  mark  of  a  Christian, 
and  I  think  there  was  often  some  doubt  of  our  really 
belonging  to  that  religion,  since  we  did  not  wear  it.  I 
met  one  enthusiastic  chaplain,  who  made  a  public 
profession  of  his  faith  by  conspicuously  wearing  a  blue 
necktie. 

Near  Agula  I  at  last  found  some  faii-ly  preserved 
fossils  in  the  limestone.  All  were  bivalves,  but  they 
comprised  Ceivmya,  Pholadomya,  and  Trigonia  of  typi- 
cally oolitic  forms,  which  fully  confirmed  the  opinion 
expressed  by  the  energetic  French  travellers,  MM.  Ferret 
and  Galinier,  who  first  explored  these  formations.  At 
Dongolo,  after  much  search,  I  obtained  several  well- 
preserv^ed  specimens  of  a  small  Ilemicidarisy  also  Avith 
characteristically  oolitic  afl&nities. 

Instead  of  following  the  road  from  Agula  to  Dongolo, 
I  went  for  some  miles  down  the  valley  of  the  Agula 
river,  which  runs  in  a  deep  dell  between  cliffs  of  lime- 
stone ;  and  then,  crossing  the  hills  to  the  Dongolo  stream, 

h2 


100  PERSONAL  NARRATlFZ 

which  flows  in  a  similar  deep  ravine,  I  returned  by  it  to 
the  road,  Sir  C.  Staveley  having  allowed  me  to  take  a  small 
guard.  I  was  unable,  however,  to  reach  sufficiently  far 
to  examine  the  base  of  the  limestone  in  this  direction.^ 

At  Dongolo  I  halted  for  a  day,  and  was  overtaken  by 
the  Commander-in-chiefs  camp  and  the  rear-guard.  It 
had  been  the  intention  of  Sir  Robert  Napier  that  a  small 
party  with  a  guard  should  be  allowed  to  cross  from  the 
Haramat  plain  near  Adabagi  to  Adowa  and  Axum,  and 
thence  march  to  rejoin  the  main  body  at  Senafe  ;  and  to 
this  party  Dr.  Cook,  Mr.  Holmes  the  archaeologist,  the 
officers  of  the  Trigonometrical  Survey,  and  a  few  others, 
including  myself,  were  to  have  been  attached.  I  was 
especially  desirous  of  examining  the  remarkable  Adowa 
hills,  reported  by  some  travellers  to  be  of  sandstone,  but 
which  Dr.  Schimper  considers  of  volcanic  origin,  and 
which  are,  as  previously  remarked,  probably  identical  in 
composition  with  the  trachytic  rocks  about  Senafe,  which 
they  resemble  so  greatly  in  form.  However,  it  was 
found  that  the  length  of  the  march  was  too  great  to  enable 
the  party,  if  detached,  to  rejoin  the  main  body  before  the 
latter  left  Senafe,  so  the  project  was  abandoned. 

I  returned  with  the  head-quarters  camp  to  Adigrat  and 
Senafe.  At  the  former  place  I  found  that  the  collector  I 
had  left  behind  had  not  a  large  number  of  bird-skins, 
though  there  were  many  amongst  them  of  interest.  He 
had,  however,  about  twenty   specimens   of  the   species 

1  Amongst  a  collection  of  rock  specimens  which  Dr.  Beke  has  veiy  politely 
sent  to  me  for  examination,  I  find  fragments  of  both  limestone  and  sandstone 
from  the  country  farther  to  the  westward  in  this  direction. 


SENAFK  •'"ID  I 

'  *  •    " 

of  hyrax   found   in   the   neighbourhood,    which   serv^'- 
admirably  to  illustrate  the  astonishing  variation  in  the..- 
colour  and  even  the  texture  of  the  fur  which  is  exhibited    , 
by  these  animals,  and  which  is  scarcely  equalled,  so  far  as 
I  know,  by  any  other  wild  mammal. 

We  reached  Senafe  on  the  25th  May,  Very  little  rain 
had  fallen  in  this  part  of  the  country,  and  it  was  conse- 
quently less  green  than  in  February,  and  contrasted 
strongly  with  the  southern  portions  of  Tigre  between 
Antalo  and  Adigrat  Kassa,  the  Tigre  chief,  was  en- 
camped in  the  neighbourhood,  and  durbars  and  military 
exercises  by  both  Abyssinians  and  British  troops  were  the 
order  of  the  day. 

At  Senafe  I  first  met  Mr.  Jesse,,  the  gentleman  sent  out 
1^  the  Zoological  Society  of  London;  He  had  hitherto 
been  most  unfortunate,  for  he  bad  only  landed  in  March, 
and  had  been  detained  in  the  rear  both  by  want  of 
carriage  and  by  illness.  His  collections  at  this  period 
were  consequently  comparatively  small  The  collector 
whom  I  had  left  behind  at  Undul  Wells  had  proved 
thoroughly  worthless,  and  had  found  it  more  profitable  to 
skin  birds  for  some  of  the  officers  than  to  do  his  own 
work.  Lieut  Sturt,  however,  had  a  considerable  number 
of  birds,  of  which  he  allowed  me  to  take  such  as  I  had 
not  myself  collected.  Altogether,  before  I  left  the  high- 
lands I  had  about  900  specimens-  of  birds,  mammals,  and 
reptiles,  representing  above  250  species. 

A  considerable  change  had  taken  place  in  the  avi- 
fauna, caused  chiefly  by  the  immigration  of  birds  from 
lower  elevations  and  the  departure  of  the  Saxicolce,  and 


•  -». 


•     • 


1  dg  •  "•  PERSONAL  NARRATIVE. 

•  • 

i^me  other  migratory  kinds.     Of  the  immigrants,  perhaps 

'*.  Jhe  most  conspicuous  was  Lamprocolius  chalybmis,  now 

abundant.       Tochis  Jlavirostris  and  Lamarius  cethio- 

'   picus,   which  in  February  and  March  did  not  ascend 

above  6,000  feet,  were  now  common  at  8,000. 

Senafd,  after  having  been  occupied  nearly  six  months, 
was  finally  abandoned  on  the  29th  May.  We  exchanged 
the  pleasant  climate  of  the  Abyssinian  highlands  for  heat 
and  dust,  a  short  supply  of  water,  and  universal  barren- 
ness. As  we  descended  the  pass,  every  day  brought  with 
it  additional  heat  and  discomfort,  mitigated  slightly  by 
such  European  luxuries  as  preserved  provisions,  wine, 
and  beer,  which  had  been  imported  in  abundance  by 
enterprising  speculators. 

In  the  pass  I  found  that  most  lovely  bird,  Pholidauges 
Uiicogaster^  abundant  from  about  2,000  to  5,000  feet 
of  elevation.  It  had  certainly  been  entii-ely  wanting  in 
January  and  February ;  so  conspicuous  a  bird  could  not 
possibly  have  escaped  every  one's  notice.  At  Komayli, 
at  the  base  of  the  hills,  I  found  some  other  birds, 
especially  (Eiia  capensistmdi  Nectannia  metallica,  which 
were  certainly  not  to  be  found  there  in  February.  Both 
were  now  breeding.  A  bee-eater,  Merops  alhicoUis,  had 
also  been  added  to  the  fauna ;  whilst  Cercomela  mela- 
nura,  Aedon  galactodes,  and  Turtur  dlbiventris  appeared 
much  more  abundant  than  formerly.  On  the  other  hand, 
all  the  shrikes,  SaxicolcB,  and  wagtails,  with  Corpus 
affinisj  had  left. 

I  remained  a  few  days  in  Komayli.  The  heat  was 
excessive ;  the  temperature  at  mid-day,  in  a  large  tent 


ZULLA.  103 

with  a  double  roof,  affording  almost  as  perfect  protection 
as  a  thatched  house,  bemg  11 4°  to  115**  Fahr.  In  the 
afternoon  a  strong  breeze  set  in  from  the  sea,  which 
slightly  abated  the  temperature.  I  returned  to  ZuUa  on 
the  6th  June,  and  took  up  my  old  quarters  in  the  Com- 
missariat enclosure,  one  of  the  best  parts  of  the  camp, 
for,  being  close  to  the  sea,  it  was  much  less  dusty  than 
further  inland. 

Sir  Eobert  Napier  left  on  the  9th  for  England,  and 
only  a  small  portion  of  the  army  remained  in  charge  of 
the  stores  which  were  being  shipped.  I  determined  to 
see  the  last  of  the  expedition.  Six  months'  campaigning 
had  left  me  in  the  best  possible  health,  and  I  only 
regretted  being  obliged  to  return  to  a  more  civilized 
country.  I  had  arranged  to  go  back  to  India  by  one  of 
the  steamers  which  was  to  remain  at  Annesley  Bay  till 
the  last,  when  an  opportunity  offered  itself  of  seeing 
something  more  of  Abyssinia. 

Lieut.  Mockler  had  been  one  of  the  first  oflScers  of 
the  army  who  landed  in  Abyssinia  as  an  assistant  in 
the  Political  Department.  During  the  advance  of  the 
army  on  Magdala,  he  had  remained  on  the  coast  in 
charge  of  the  very  important  political  relations  with  the 
Shoho  and  other  chieftains  which  had  been  so  ably 
originated  by  Colonel  Merewether  and  Mr.  Munzinger. 
On  the  departure  of  the  force  he  had  applied  for  leave 
of  absence  in  order  to  visit  the  Anseba  and  Bogos 
country,  a  friendly  territory  lying  about  one  hundred 
miles  north-west  of  Massowa.  Mr.  Munzinger,  who  has 
passed  many  years  in  Bogos,   also  intended  to  spend 


104  PERSONAL  NARRATIVE. 

part  of  the  summer  there ;  and  when  Lieut  Mockler 
asked  me  to  join  the  party,  the  temptation  was  too  great 
to  be  resisted.  Mr.  Jesse  the  zoologist,  who,  like  myself, 
was  eager  to  see  something  more  of  the  country  and  it3 
fauna  before  leaving  it,  joined  us  ;  and  harvirig  purchased 
from  the  Commissariat,  which  was  v6ry  glad  to  dispose 
of  them,  flour,  rice,  and  grain  for  ourselves  and  the  men 
whom  we  intended  to  take  with  us,  sufficient  for  a 
couple  of  months,  we  left  ZuUa  on  the  18th  June,  when 
it  was  finally  abandoned  by  the  army,  and  went  to 
Massowa  in  a  steamboat  which  was  despatched  to  carry 
Mr.  Munzinger  thither. 

Massowa  has  been  described  by  almost  every  writer 
who  has  visited  Abyssinia.  It  is  a  small  Arab  town  on 
an  island  about  a  quarter  of  a  mile  from  the  shore,  and 
without  any  water.  It  is  probably  one  of  the  hottest 
places  in  the  Vrorld,  and  the  change  from  Annesley  Bay 
was  far  from  agreeable,  although  we  had  a  house,  the 
British  Consulate,  to  live  in,  a  luxury  to  which  we  had 
all  been  strangers  for  many  months.  I  believe  we  all 
longed  to  be  in  tents  again ;  I  did  certainly,  but  three 
or  four  days'  delay  was  necessary  in  order  to  make 
arrangements  for  carriage.  A  large  number  of  camels 
had  been  left  behind  by  the  Transport  Train ;  we  obtained 
the  loan  of  some,  but  as  they  had  no  saddles  or  ropes 
for  loading,  we  w^ere  obliged  to  procure  these  also. 

All  being  ready,  we  left  Massowa  on  the  afternoon  ol 
the  22d  of  June,  and,  crossing  the  arm  of  the  sea  whic^! 
forms  the  harbour,  rode  for  about  four  miles  inland  tA 
MakuUu,  a  large  village  whence  a  gre«qt  portion  of  the 


Mjssojrj,  105 

water  used  at  Massowa  is  brought,  and  where  most  of 
the  Indian  and  Arab  merchants  have  houses  and  live 
during  the  hot  months,  going  to  Massowa  in  the  day 
only. 

The  spur  of  land  opposite  Massowa,  and  the  island 
itself,  consist  of  the  coral  rock  of  which  so  many  of  the 
islands  in  this  part  of  the  Red  Sea  are  composed.  It  is 
formed  by  the  agglutination  of  broken  corals,  shells,  and 
coral  sand.  After  leaving  the  shore  a  flat  alluvial  plain 
is  traversed  as  far  as  Makuilu.  Here  hills  formed  of 
volcanic  rocks  appear,  similar  to  those  near  Zulla. 

At  Makuilu  we  took  leave  for  the  time  of  Mr.  Mun- 
zinger,  who  was  obliged  to  remain  in  Massowa  for  some 
days,  in  order  to  make  arrangements  for  the  care  of  the 
transport  animals  left  in  his  charge.  We  started  early 
the  next  morning,  and  marched  about  twelve  miles  due 
west  to  Saati,  through  low  hills  of  volcanic  and  sedi- 
mentary rocks  intermixed,  all  apparently  belonging  to 
the  same  group,  and  having  a  general  dip  to  the  east. 
Basaltic  lava  predominates.  The  ground  is  mostly 
rather  barren,  stunted  mimosa  bushes  of  the  most  thorny 
description  being  almost  the  only  vegetation. 

At  Saati  there  is  a  spring  of  water,  and  we  halted 
here  for  the  night.  I  was  rather  unwell,  having  been 
knocked  up  by  the  heat  of  Massowa,  and  lame  from 
some  unknown  cause,  so  I  could  not  go  out.  In  the 
evening  towards  dusk  thousands  of  sand-grouse  assem- 
bled at  the  water  to  drink,  coming  from  all  quarters. 
All  appeared  to  be  of  one  kind,  Pterocles  lAchteiisteiniy 
which  is  common  throughout  this  part  of  the  country. 


106  PERSONAL  NARRATLVR 

In  other  places  we  subsequently  often  saw  large  num- 
bers of  this  bird  flying  to  water  in  the  evening  and 
early  morning,  but  nowhere  so  many  as  at  Saati.  All 
sand-grouse  have  particular  hours  of  the  day  for  drinking, 
and  PL  Lichteiisteiniy  like  the  closely-allied  Pt.  fcisci- 
atus  of  Hindustan,  drinks  at  twilight  in  the  morning 
and  evening. 

In  the  morning  I  found  large  numbers  of  bunting 
{Eniberiza  septemstriata)  at  the  water.  We  started  at 
daybreak,  and  marched  west  for  about  fifteen  miles  to 
Ailat  A  small  outcrop  of  granitic  rocks  appears  at 
Saati  from  beneath  the  volcanic  formations,  and  the 
spring  is  probably  brought  to  the  surface  in  consequence 
of  their  existence,  the  water  in  the  porous  scoriaceous 
rock  being  dammed  back  and  forced  to  the  surface  by 
the  impervious  granitoid  gneiss.  The  latter  formation, 
doubtless,  underlies  the  volcanic  beds  throughout,  and 
probably  at  no  great  depth.  After  passing  Saati,  volcanic 
rocks,  precisely  similar  to  those  further  east,  are  met 
with  for  about  three  or  four  miles,  when  metamorphics 
succeed  them;  the  road  traverses  low  metamorphic 
hills  for  a  few  miles  further,  and  then  emerges  upon 
a  plain  four  or  five  miles  broad,  in  which  is  the  vil- 
lage of  Ailat,  consisting  of  thatched  huts  surrounded 
by  high  thorn  fences  to  keep  out  wild  beasts,  here 
numerous. 

The  plain  around  Ailat  was  tolerably  green,  even"  at 
this  period  of  the  year;  large  tamarisk  trees  grew  in 
places  along  a  broad  sandy  watercourse  which  traversed 
it ;  and  they,  with  a  few  other  plants,  gave  the  country 


JILAT- LIONS.  107 

an  appeaxance  of  fertility  which  contrasted  agreeably 
with  the  barren  rocks  between  which  we  had  l)een 
travelling.  We  crossed  the  plain,  and  went  on  al)out 
three  miles  beyond  Ailat  to  encamp  near  the  famous  hot 
spring.  The  village,  as  is  usual  tliroughout  Northern 
Abyssinia,  is  at  a  distance  of  two  or  three  miles  from 
water ;  the  cause  assigned  by  the  people  for  this  singular 
practice  is,  that  plundering  bands  often  encamp  near  the 
water.  Probably  the  danger  from  wild  beasts,  whicli 
come  down  to  drink  in  considerable  numbers,  is  another 
reason  for  not  living  close  to  the  springs.  Of  this  risk 
wo  had  a  most  melancholy  practical  illustration  a  few 
days  later. 

The  hot  spring  at  Ailat  issues  from  some  rocks  in  the 
bed  of  a  torrent,  which  is  dry  except  after  heavy  rain 
in  the  hills.  The  water  is  extremely  hot,  140°  Fahr., 
but  perfectly  tasteless. 

I  did  not  go  out  next  day  until  the  evening,  when 
some  people  saw  a  lioness  close  to  our  camp.  Mockler 
and  I  went  to  look  for  her,  and  after  peering  about  some 
time,  Mockler  saw  her  in  a  bush,  and  fired,  but  missed. 
I  could  not  see  the  animal  distinctly.  She  got  amongst 
some  dense  green  bushes,  and,  as  it  was  already  dark, 
we  were  obliged  to  leave  her. 

The  next  morning  wt.  took  up  her  tracks  ;  she  had 
been  joined  by  a  lion  in  the  night,  and  had  gone  to  the 
north.  We  did  not  succeed  in  finding  her,  however.  I 
killed  a  large  pig  {Phacochceims),  of  which  I  preserved 
the  skull,  and  shot  several  birds,  amongst  which  were 
Lamprocolius  chrysaffoster,  Halcyon  semiccevulea,  and 


108  PERSONAL  NARRATIVE, 

GentropiLs  superciliosuSy  none  of  which  I  had  previously 
seen.  In  the  evening  the  lioness  again  crossed  the  stream 
near  out  camp,  and  we  follot\red  her,  but  she  went  away 
at  once.  In  going  after  her  we  siaw  a  leopard,  which 
also  escaped. 

Ailat  is  notorious  for  lions  and  leopards,  many  of 
which  are  man-eaters,  and  many  people  are  killed  by 
these  animals.  It  is  a  most  curious  circumstance  that 
whereas  on  the  route  followed  by  the  army,  not  forty 
miles  farther  south,  lions  were  only  seen  on  two  or  three 
occasions,  and  not  a  single  animal  was  shot,  although 
there  were  dozens  of  sportsmen  eager  for  the  chance, 
at  Ailat,  and  througliout  our  subsequent  journey,  we 
frequently  met  them  ;  and  scarcely  a  day  passed  without 
our  seeing  their  fresh  tracks. 

The  temperature  was  certainly  very  hot,  rising  to 
about  108°  in  the  day,  and  falling  to  75°  in  the  early 
morning  before  sunrise ;  but  this  was  an  agreeable  change 
from  Massowa,  where  night  and  day  the  thermometer 
never  fell  below  about  87°  or  88°  Fahr.  I  very  soon 
recovered,  but  Jesse  unfortunately  had  a  rather  smart 
attack  of  fever,  apparently  produced  by  exposure  to 
the  sun. 

I  procured  many  birds  during  the  six  days  we  re- 
mained at  Ailat ;  amongst  them  were  Nisus  niloticxcs, 
Nilaus  brubru,  Ibckics  erythrorhynchuSy  Quelea  cethi- 
opica,  Peristera  afra,  and  JEgiatilis  tricollwns.  I  also 
found  some  species  which  elsewhere  I  had  only  met 
at  a  much  greater  elevation  (Ailat  is  less  than  1,000 
feet  above   the   sea),  such   as   Tchitrea  melanogastra. 


A  MAN  KILLED  BY  A  LEOPARD,  109 

Dicrurus  divaricatus,  Lamprocolius  chalybcBus,  Buphaga 
erythrorhyncftAXy  and  especially  Irrisor  erythrorhynchus, 
Passer  Swamsoni,  and  Hophpterus  spinosus.  Bus- 
tards {Otis  arahs),  guinea-fowl,  and  francolins  (Pter- 
nestes  ruhricollis)  abounded. 

On  the  early  morning  of  the  29th  June  one  of  my 
servants,  rising  before  daybreak,  was  scratched  in  the 
face  by  some  wild  animal  which  had  come  into  the  camp. 
The  track  resembled  that  of  a  large  cat.     We  thought 
nothing  of    this   at    the    time,   but  on   the   following 
night  we  were  all  aroused  by  an  outcry  and  shouting, 
and   an   alarm  was  given  that  a  lion  had  seized   one 
of  our  men.      Mockler  fired   off  his  rifle   to   frighten 
away  the   beast,  which  rushed   roaring  past  our  tent. . 
On  inquiry  we  were  horrified  to  find  that  an  Abyssinian 
servant  of  Jesse's  had  been  killed  while  asleep,  and  no 
alarm  had  been  created  until  the  animal  attempted  to 
drag  away  the  body.     The  unfortunate  man  had  two 
large  tooth-holes  in   his   throat,  and  must  either  have 
been  so  seized  that  he  was  unable  to  cry  out,  or  else, 
as  is  probable,  his  neck  was  broken.     The  assailant  was 
doubtless  a  leopard,  very  probably  the  same  small  animal 
which  had  scratched  my  servant  the  night  before.     A 
lion  would  have  carried  away  the  body  to  some  distance, 
certainly,  and  tlie  tooth-marks  were  too  small  to  have 
been  made  by  a  lion's  fangs.     We  had  a  low  thorn  fence 
round  three  sides  of  our  camp,  and  the  camels  occupied 
the  open  side — tlie  usual  plan  in  this  part  of  Africa — 
but  w-e  had  no  fires,  a  most  necessary  precaution,  and 
one  we  never  neglected  after  this  sad  lesson. 


110  PERSONAL  NARRATIVE. 

During  all  my  experience  in  India — and  in  the  course 
of  twelve  years  I  have  encamped  in  many  places  infested 
by  tigei*s  and  leopards — I  have  never  taken  such  pre- 
<3autions  as  we  had  adopted  even  before  this  accident. 
Throughout  the  progress  of  the  expedition  in  Abyssinia, 
no  one  ever  dreamed  of  the  necessity  for  especial  pro- 
tection against  wild  animals. 

In  the  morning,  all  attempts  to  track  the  leopard 
failed ;  we  could  not  find  a  single  distinct  footprint. 
The  poor  fellow  who  had  been  killed  was  buried,  and 
a  pile  of  stones  heaped  over  his  grave.  To  have  re- 
mained at  Ailat  would  simply  have  been  to  risk  a 
repetition  of  the  disaster ;  consequently,  after  expending 
the  morning  in  a  vain  attempt  to  discover  the  beast  and 
avenge  the  death  of  our  poor  follower,  we  packed  up 
imd  started  for  Asus,  a  village  about  eight  miles  to  the 
north,  of  very  similar  appeai'ance  to  Ailat,  and  lying  in 
the  same  plain,  which  extends  here  for  many  miles  along 
the  base  of  the  lowest  hills. 

The  ranges  in  this  neighbourhood  rise  far  more  gently 
from  the  low  country  than  they  do  near  Annesley  Bay. 
The  high  plateau  of  Hamazen,  equal  in  elevation  to  the 
country  around  Senaf6  and  Halai,  is  visible  to  the  west ; 
but  the  intermediate  ground,  instead  of  consisting  of  deep, 
rugged  ravines,  is  composed  of  rounded  hills,  drained  by 
valleys  of  no  great  depth. 

We  halted  a  day  at  Asus,  and  then  went  into  the 
hills  again  to  Kusaret,  on  the  road  to  Mensa,  intend- 
ing to  go  to  a  place  called  Tunfia,  said  to  be  much 
o.ooler  than  Ailat,  and  situated  at  an  elevation  of  about 


MARCH  TO  NORTHWARD,  \\\ 

2,000  feet.  But  at  this  time  wc  heard  from  Mr.  Mun- 
zinger  that  he  expected  still  to  be  detained  for  several 
days  in  Massowa,  and  that  we  had  better  proceed  to 
Bogos,  by  Ain  and  the  Lebka  valley,  without  waiting  for 
him.  One  of  the  Naibs  of  Arkiko  was  with  us,  and  his 
presence  would  be  ample  guarantee ;  in  fact,  the  people 
being  friendly,  no  special  protection  was  necessary. 

We  were  all  tired  of  the  heat,  and  Jesse  was  for  a  day 
or  two  veiy  ill.  We  therefore  determined  to  proceed  at 
once  to  Ain,  and  thence  to  a  cooler  climate.  We  accord- 
ingly retraced  our  steps  somewhat,  and  then  marched  to 
Kanzal,  still  in  the  plain  i-unning  northward  from  Ailat ; 
but  at  Kanzal  the  hills  to  the  east  become  scattered,  and 
hence,  to  the  northward,  one  great  barren  plain  extends 
from  the  base  of  the  mountains  to  the  sea.  This  tract  is 
a  perfect  desert  in  the  summer,  though  after  the  winter 
rains  it  becomes  green,  and  affords  pasture  to  the  cattle 
of  the  Habab  tribes.  The  heat  and  glare  in  the  day- 
time at  the  season  of  our  journey  were  said  to  be  almost 
unbearable,  and  it  was  therefore  necessary  to  make  the 
march  across  the  desert  of  Shob  (or  Sha  ab),  as  this  tract 
is  called,  at  night.  We  started  in  the  evening  of  the 
4th  of  July,  with  a  full  moon. 

Before  dark  we  passed  a  large  encampment  of  the 
Warea,  one  of  the  nomade  Bedouin  tribes.  It  was  sur- 
rounded by  a  low  circular  thorn  fence,  immediately 
inside  which  was  a  ring  of  hemispherical  huts,  the  frame- 
work of  bent  sticks,  covered  by  mats.  These  huts  can 
be  taken  to  pieces,  and  packed  on  a  camel  or  a  bullock, 
and  set  up  afresh  in  a  very  short  time.     We  found  them 


112  PERSOyjL  XARR.ITIFE, 

generally  used,  not  only  throughout  the  nomade  Mahom- 
medan  tribes  of  the  Habab,  but  also  amongst  the  agri- 
cultural Christian  population  of  the  Anseba  valley  and 
Bogos.  Even  when  thatched  huts  are  used,  as  at  Keren, 
the  hemispherical  mat  dome  is  placed  inside. 

In  the  centre  of  the  circle  were  two  or  three  isolated 
huts,  said  to  be  used,  one  for  marriages,  and  another  for 
the  sick  I  doubt  if  these  be  really  their  use.  The 
inhabitants  were  much  like  Shohos,  but  a  finer  race.  The 
men  had  enormous  masses  of  frizzled  hair,  in  which  were 
stuck  the  usual  two  or  three  long  skewers  of  wood  or 
horn.  Like  all  the  people  in  Northern  Abyssinia,  they 
wore  large  straight  swords,  of  German  manufacture, 
instead  of  the  little  sickle-like  weapons  of  the  Shohos,  or 
the  longer  curved  scimitars  of  the  Tigr^ans  and  other 
Abyssinians. 

Uncouth  as  they  appeared,  these  people  were  \Qvy 
civil,  and  their  chief  went  on  with  us  for  some  distance 
to  point  out  the  road.  As  w^e  came  out  upon  the  open 
desert  we  saw  many  large  gazelles  {G.  Sceinmeriiigii)  in 
the  moonlight ;  but  they  would  not  allow  us  to  approach 
sufficiently  close  to  shoot 

We  reached  Ain  in  the  morning.  A  small  stream,  fed 
by  a  spring,  runs  between  thick  green  bushes  and  reeds 
in  the  bed  of  the  Lebka  torrent.  Here  we  halted  for 
a  day.  The  slope  of  the  coast  plain  must  be  considerable, 
since  at  Ain,  about  fifteen  miles  from  the  sea,  the  eleva- 
tion by  aneroid  is  about  1,200  feet.  All  the  bills  around 
are  schistose  and  gneissic  rocks,  as  are  indeed  all  between 
Ailat  and  Ain.     The  general  strike  is  still  north  and 


LEBKA  r ALLEY,  113 

south.     The   dip    at  Ain   is   low,    20"*  or  30'  to  the 
west 

There  were  not  many  birds  here  which  we  had  not 
obtained  at  Ailat,  Tockus  nasutus  and  Crateropus  leu- 
cocephalus  being  the  only  important  additions :  but  we 
were  told  that  a  herd  of  oryx  (0.  Beisa)  inhabited  the 
plain  to  the  eastwards.  Of  all  the  antelopes  of  Africa 
none  is,  perhaps,  so  remarkable  as  the  oryx,  and  I  longed 
to  see  one.  But,  although  we  found  fresh  tracks,  we 
could  not  meet  with  the  animals  themselves. 

On  the  7th  we  marched  up  the  Lebka  valley  to 
Mohabar,  a  distance  of  about  twenty  miles.  The  ascent 
is  very  gradual,  and  the  valley  does  not  contract  as  in 
the  Komayli  pass.  As  a  rule  the  road  is  very  easy  along 
the  broad  sandy  bed  of  the  Lebka,  which  early  in  July 
was  quite  dry,  the  rains  in  the  higher  part  of  the  valley 
having  scarcely  commenced.  In  one  or  two  places  the 
ravine  becomes  narrower,  with  steep  sides  ;  one  of  these 
places  is  known  as  Aualid  Oret  ("the  Daughters  of 
Hades"),  but  its  appearance  is  less  formidable  than  its 
name.  The  bed  of  the  stream  is  rocky,  but  our  loaded 
camels  got  over  without  much  difficulty. 

We  encamped  in  the  bed  of  the  stream.  The  next 
morning  we  found  that  a  large  lion  had  walked  past  our 
camp  and  down  the  stream  during  the  night.  We  took 
up  his  tracks,  but  unsuccessfully.  The  day  before  a  lion, 
probably  the  same,  had  killed  a  cow  and  then  a  man 
close  to  Mohabar. 

The  sandy  ravine  bed  was  fringed  here  with  fine 
tamarisk  trees,  which  abounded  in  hawks  and  falcons. 


114,  PERSONAL  NARRATIVE. 

The  little  Ntsus  sphenurus  was  especially  abundant, 
feeding  on  cicadcB.  It  is  a  smaller  bird  than  the  closely 
allied  shikra  of  India  {Nisits  hadius)^  and  more  insec- 
tivorous. 

At  night  it  was  cloudy,  a  little  rain  fell,  and  we 
moved  our  tents  and  baggage  hurriedly,  for  there  was 
lightning  and  thunder  to  the  westward  at  no  great  dis- 
tance, and  our  people  feared  that  the  stream  might  come 
down  suddenly.  We  saw  some  of  the  sudden  floods  so 
prevalent  in  these  torrent  beds  afterwards,  and  certainly 
they  are  to  be  feared  However,  on  this  occasion  the 
water  did  not  reach  us. 

Another  long  march  up  the  valley  produced  a  great 
change.  As  far  as  Kelamet,  twelve  miles  from  Mohabar, 
we  were  still  amongst  parched  and  barren  hills,  but  at 
an  elevation  of  about  3,500  feet  we  came  almost  sud- 
denly into  a  tract  of  hiUy  country  covered  with  the 
richest  verdure.  Green  bushes  and  kolqual  trees  covered 
the  hill-sides,  magnificent  baobab  trees  were  dotted  over 
the  valleys,  every  little  glade  was  covered  with  green 
grass.  We  had  passed  from  the  coast  region,  in  which 
no  rain  falls  during  the  summer  months,  into  the  high- 
lands, in  which  the  monsoon  rains  last  from  June  till 
October,  as  in  South-eastern  Asia. 

The  temperature  was  delicious.  Life  seemed  to  return 
with  new  zest  A  day  or  two  sufficed  to  banish  all  the 
effects  of  the  heat  at  Massowa  and  Ailat.  The  animals 
seemed  refreshed  like  the  plants,  and  birds  were  singing 
in  all  directions.  Many  of  these  were  quite  new  to  me  ; 
a  parrot,  Pionus  Meyen\  was  common,  and  no  less  than 


KOKAI^WILD  ELEPHANTS.  115 

three  kinds  of  rollers,  Coracias  ahyssinica,  C.  pilosa, 
and  Eurystomus  afer.  The  first  I  had  found  in  places 
on  the  highlands,  the  second  I  had  once  or  twice  seen 
only,  the  third  I  had  never  previously  met  with.  Here 
all  three  were  common. 

At  Kokai,  where  we  pitched  our  tents,  there  was  a 
small  encampment  of  the  Habab,  and  a  very  large  herd 
of  camels.  Many  cattle  had  been  brought  into  the 
neighbourhood  for  pasture.  The  wild  animals  migrate 
from  the  lowlands  at  this  season  like  the  men,  and  we 
heard  of  wild  elephants  as  soon  as  we  arrived. 

The  next  day,  July  1 0th,  some  men  brought  us  news 
of  a  herd  in  the  neighbourhood,  and  on  going  out  we 
saw  an  elephant,  to  our  utter  astonishment,  standing 
amongst  high  bushes,  with  camels  scattered  around  in 
all  directions.  Some  other  elephants,  it  was  uncertain 
how  many,  were  in  higher  jimgle  close  by.  The  camels 
had  been  grazing  in  the  neighbourhood  all  the  morning, 
and  the  drivers  had  been  shouting  to  each  other  and  to 
their  animals.  Certainly  the  elephants  cared  wonder- 
fully little  for  the  presence  of  domestic  animals,  or 
even  of  men. 

We  lay  still  for  some  time  and  watched  the  huge 
beast,  which  stood  about  200  yards  away,  lazily  munch- 
ing leaves,  or  flapping  its  large  ears.  There  was  no 
chance  of  approaching  close,  but  by  making  a  circuit  we 
reached  within  about  sixty  yards  without  being  seen, 
and  Mockler  fired  a  shell,  which  evidently  hit,  and  the 
elephant  dashed  into  the  jungle  in  front,  issuing  again  in 
front  of  us  immediately  after,  followed  by  two  other  fuU- 

I  2 


116  PERSONA!.  NARRATIFK 

grown  elephants  and  two  young  ones.  As  they  crossed 
in  front  of  us  at  about  forty  yards  Jesse  and  Mockler 
both  fired  at  the  leading  elephant,  which  fell  over.  The 
others  stood  by  it,  and  were  soon  dropped  also ;  but  we 
had  to  fire  several  bullets  before  they  were  dead.  Not  a 
single  attempt  at  a  charge  was  made  by  any  one  of  them. 
The  smallest  elephant  was  shot  by  accident. 

A  more  unexciting  attack  it  would  be  impossible  to 
conceive,  and  I  believe  we  all  felt  ra|;her  ashamed  of  our- 
selves for  shooting  such  inofiensive  animals  as  these 
elephants  appeared.  Abyssinian  elephants  are  not  always 
80  harmless.  An  officer  of  the  2d  Belooch  regiment  was 
severely  injured,  and  very  nearly  killed  by  one  which  he 
had  wounded  near  Suru.  Lieut.  Mockler  was  most 
viciously  pursued  by  a  herd  which  he  attacked  on  the 
Anseba ;  and  Mr.  Jesse  was  charged  by  a  large  female 
without  any  provocation,  on  our  return  journey,  close 
to  this  very  place,  Kokai.  For  my  own  part,  this 
happened  to  be  the  first  and  last  occasion  on  which 
I  came  across  any  elephants  in  Abyssinia.  The  only 
satisfaction  was  that  we  provided  all  the  people  in  the 
neighbourhood  with  meat  for  some  time.  We  made 
one  or  two  essays  ourselves.  Trunk  proved  rather 
good,  though  tough ;  but  elephant  foot  met  with  general 
disapproval 

On  the  12th,  Mockler  went  on.  Jesse  and  I  remained 
a  day  longer  to  collect  at  Kokai,  and  proceeded  on  the 
13th.  The  road  led  through  fine  jungle  for  about  five 
miles,  gradually  ascending.  All  the  rocks,  as  throughout 
the  Lebka  valley,  are  metamorphic,  but  from  Mohabar 


ARRIFAL  Ili  THE  ANSEBA  FALLEF.  117 

there  is  a  considerable  change  in  the  foliation,  which  is 
here  north-east  and  south-west. 

After  about  two  hours'  gradual  ascent,  we  came  to  a 
steeper  path  leading  over  a  small  range  of  hills  between 
the  upper  portion  of  the  Lebka  valley  and  the  basin  of 
the  Anseba.  From  the  top  of  the  little  pass,  which 
is  called  Mashalit,  there  is  a  jBne  view  over  the  broad 
undulating  valley  in  which  the  Anseba  runs.  We  de- 
scended a  finely-wooded  valley,  and  found  ourselves  in 
an  open  couiitry,  with  more  signs  of  inhabitants  than 
the  Lebka  valley,  and  cultivated  in  places.  The  popu- 
lation; however,  owing  to  Abyssinian  raids  and  inter- 
necine disputes  and  blood-feuds,  is  said  to  have  much 
diminished  of  late  years,  and  the  area  actually  under 
cultivation  is  small  compared  with  that  which  has 
evidently  been  cleared. 

The  most  remarkable  feature  of  the  valley  is  the 
existence,  on  almost  every  conspicuous  rise,  of  large 
circular  piles  of  white  stones.  Nothing  of  the  kind  had 
been  seen  by  us  in  the  Abyssinian  highlands.  These 
piles  of  stones  cover  the  graves  of  chiefs.  In  some  cases 
the  piles  are  not  covered  with  the  white  fragments  of 
quartz  which  are  usually  placed  on  the  surface,  and  to 
which  these  sepulchral  moimds  owe  their  conspicuous 
appearance.  Such  graves,  we  were  told,  are  those  of  men 
who  have  been  murdered,  and  whose  deaths  have  not 
been  avenged. 

The  law  of  blood  for  blood,  universal  amongst  Semitic 
peoples,  prevails  generally  throughout  Abyssinia.  It  is 
of  this  nature   in  the  Anseba  country.     If  a  man  is 


118  PERSONAL  NARRATIVE. 

murdered  or  slain  accidentally,  it  matters  not  how,  his 
family  may  agree  to  receive  blood-money  from  the 
slayer,  the  price  being  fixed  usually  at  a  certain  number 
of  cows,  difiering  according  to  his  status  as  chief  or  serf ; 
but  should  the  relations  refuse  to  accept  compensation, 
or,  as  is  commonly  the  case,  should  the  murderer  and  his 
family  be  unwilling  or  unable  to  pay,  a  blood-feud  arises, 
and  the  relations  of  the  dead  man,  up  to  the  seventh 
degree,  are  bound  to  kill  either  the  murderer  or  one  of 
his  near  male  relations,  also  to  the  seventh  degree.  As 
may  be  easily  supposed,  the  feud  either  between  families 
or  tribes  is  seldom  at  an  end  with  the  loss  of  one  or  two 
lives  :  usually  it  continues  until  one  or  the  other,  by  re- 
peated losses,  has  become  so  weakened  that  it  sues  for 
peace,  when  an  agreement  is  made,  and  sealed  by  inter- 
marriage; a  girl  from  each  family  or  tribe  being  given  in 
marriage  to  a  man  of  the  other.^ 

In  many  respects  the  people  of  the  Anseba  are  a  most 
interesting  race,  or  grotip  of  races.  Unlike  the  Habab 
and  Samhar  tribes  inhabiting  the  country  to  the  east  and 
north-east,  they  are  still  for  the  most  part  nominally 
Christians.  Forty  years  ago,  all  the  Bedouins  of  Habab 
and  Samhar  were  the  same;  but  the  Mahommedan 
religion  in  these  lands  has  progressed,  and  is  still  pro- 
gressing rapidly,  and  Mr.  Munzinger  pointed  out  to  us 

1  It  was  far  from  generally  known  in  the  army  that  poor  Dofbon's  murder 
was  in  satisfaction  of  a  blood-fend,  and  in  revenge  for  the  accidental  death  of 
a  Shoho  near  SenafS.  It  is  yeiy  much  to  be  regretted  that  the  mnrderers 
have  hitherto  been  allowed  to  remain  unpunished,  if  only  because  such 
impunity  induces  savages  like  the  Shohos  to  believe  that  Englishmen  care 
nothing  for  their  own  countrymen. 


INHABITANTS  OF  THE  AN  SEE  A  VALLEY.  119 

several  chiefs  who  had  become  converts  to  Islam.  Both 
Christianity  and  Mahommedanism  are  of  a  most  debased 
kind :  there  is  neither  church  nor  mosque,  priest  nor  mulla ; 
but  the  faith  of  the  Mussulman  is  kept  alive  by  wander- 
ing devotees,  while  the  Christians  have  only  the  Frank 
missionary,  who  preaches  to  them  a  religion  far  more 
distinct  from  their  own,  in  practice  if  not  in  creed,  than 
that  of  the  followers  of  the  Prophet 

With  the  progress  of  Mahommedanism  the  rule  of 
the  Egyptians  extends,  and  that  of  the  Abyssinians 
diminishes.  The  former  do  but  little  for  the  country, 
except  receive  taxes ;  the  latter  do  still  less,  always 
with  the  same  important  exception.  Protection  for  life 
or  property  there  is  but  little  in  Habab  or  Samhar, 
and  none  in  Tigrd. 

At  Kelamet  we  had  passed  the  last  Turkish  outpost 
The  upper  part  of  the  Lebka  valley  and  all  the  Anseba 
in  this  part  of  its  course  belongs  to  Tigrd,  and  pays 
tribute  to  the  chief  of  Hamazen.  The  tribe  owning  the 
part  of  the  valley  reached  by  us  is  that  of  Bejuk,  and 
we  rode  to  their  principal  village,  Wasentel, — generally 
known,  however,  by  the  name  of  the  tribe.  We  here 
found  Mockler  just  returned  from  exploring  the  banks  of 
the  river,  where  he  had  found  traces  of  rhinoceroses  in 
abundance.  This  was  good  news.  The  Abyssinian 
rhinoceros  had  never  been  accurately  determined,  and 
there  was  some  doubt  as  to  whether  it  was  not  a  distinct 
species  from  any  occurring  in  Southern  Africa.  On  oup 
road  we  had  met  with  three  or  four  birds  of  which  I  had 
not  specimens  before,  viz.  Nisus  niger,:  Lamprotornis 


120  PERSONAL  NARRATIFE. 

purpuropteray  Chrysococcyx  cupreus^  and  a  ruticilline 
bird,  which  I  subsequently  found  to  be  undescribed, 
mjr  jR.  fuscicaudata,  and  in  the  evening  I  added 
to  these  PalcBomis  torqucUa,  also  an  addition  to  my 
collection. 

Having  provided  my  men  with  as  many  birds  as  they 
could  possibly  skin,  we  all  started  together  the  next 
afternoon  to  look  for  rhinoceroses.  We  rode  across  the 
imdulating  plain,  cultivated  in  patches,  and  intersected 
with  ravines  filled  with  thick  bush,  to  the  river,  along 
the  banks  of  which  there  is  a  narrow  belt  of  high  forest, 
with  dense  underwood  of  bushes.  I  had  seen  nothing  so 
luxuriant  in  Africa.  So  dense  is  the  imderwood  that  it 
is  only  possible  to  creep  through  the  tangled  labyrinth  of 
roots  and  branches  by  the  paths  made  by  the  rhinoceroses 
which  haunt  these  thickets.  These  animals  retire  into 
the  thickest  parts  during  the  heat  of  the  day.  In 
particular  spots,  generally  amongst  high  reeds,  they 
select  a  small  space  of  ground,  which  they  clear  of 
bushes,  and  these  spots  are  well  known  to  the  natives  as 
"  rhinoceros  houses.*' 

In  such  a  house,  in  the  centre  of  a  broader  strip  of 
thicket  than  usual,  the  men  who  were  with  us  reported 
tliat  two  rhinoceroses,  an  old  one  and  her  cub,  were 
lying.  The  approved  method  of  endeavouring  to 
shoot  these  animals  is  by  walking  silently  into  their 
retreats,  firing  at  them,  and  then  taking  refuge  up  a 
tree.  This,  however,  requires  a  degree  of  agility  in  tree- 
climbing  upon  which  we  could  not  rely,  and  it  is  not 
only  dangerous,  but,  according  to  all  the  accounts  we 


RHINOCEROS  HUNT.  121 

rfeceived,  peculiarly  unsuccessful,  since  we  could  never 
heax  that  a  rhinoceros  had  been  bagged  by  it.  We 
accordingly  posted  ourselves  at  one  end  of  the  jungle, 
and  told  our  men  to  shout  and  make  noises  at  the  other, 
but  the  rhinoceroses  escaped  across  the  river. 

Immediately   after  this   piece   of  ill-success,  we  dis- 
covered  a  herd   of  doe  koodoo  on  the  opposite  bank. 
Riding  across  the  stream,  which  was  about  knee-deep 
with  a  rather  treacherous  bottom,  we  proceeded  to  stalk 
the   antelopes,    again   without  success,    a  man  halving 
crossed  the  ground  meantime  and  disturbed  them.     We, 
however,  caught  sight  of  them  again,  and  Mockler  shot  a 
young  doe.   As  we  came  up  to  the  body  two  rhinoceroses, 
one  a  huge  beast,  the  other  about  three  parts  grown,  ran 
across  the  open  about  150  yards  from  us,  and  disappeared 
in  a  broad  ravine.     After  peering  about  for  some  time,  a 
sharp-eyed  Somali  caught  sight  of  the  larger  one  standing 
in  the  jungle,  and  by  making  a  circuit  we  came  within 
100  yards.     Nearer  approach  was  very  diJEcult,  and  the 
bushes   were  so  high  that  if  we   entQred  the  broken 
ground,  near  the  other  side  of  which  the  animal  stood,  we 
should  probably  lose  sight  of  her  completely.     Jesse  and 
I  took  steady  aim,  and  fired.     With  the  most  astonishing 
sound,  a  series  of  snorts  like  puffs  from   an  enraged 
steam-engine,  the  brute,  with  its  young  one  following, 
dashed  into  the  bushes  in  front  and  came  on  towards  us. 
The  moment  it  left  the  spot  where  we  had  seen  it,  it 
vanished   in   the  jungle,   and  only  the   snorting  and 
crashing  of  the    branches    told  of    its  rapid  progress 
towards  us,  until  it  emerged  just  where  we  had  been 


122  "  PERSONAL  NARRATIVE. 

standing.  Meantime  we  had  modestly  retired  behind  a 
neighbouring  bush,  the  rhinoceros  continued  its  route, 
and  we  followed,  one  of  the  Bejuk  chiefs  who  was  with 
us  running  ahead.  Presently  he  rushed  back,  a  perfect 
picture  of  excitement  and  delight,  dancing  like  a  mad- 
man, beating  his  shield  with  his  spear,  and  springing 
nearly  his  own  height  into  the  air ;  telling  us,  partly  by 
words,  mainly  by  pantomime,  that  the  rhinoceros  was 
running  round  and  round  in  a  circle  and  falling  over. 
We  soon  came  up  with  it :  the  monster  lay  dying  on 
the  ground,  the  yoimg  one  stood  by  prodding  the  mother 
with  its  horns,  and  evidently  trying  to  induce  her  to 
move  on.  Though  very  badly  hit,  the  old  one  took 
several  more  bullets  before  it  was  dead;  the  smaller 
animal,  after  receiving  one  or  two  shots,  rushed  off  at 
the  most  astonishing  pace,  and  we  saw  it  no  more.  An 
examination  of  the  body  next  day  showed  that  Jesse's 
bullet^  a  conical,  had  hit  on  the  shoulder,  turned  forwards^ 
and  lodged  in  the  neck.  Mine,  a  round  bullet,  had  hit 
behind  the  shoulder,  made  a  clean  hole  through  the 
centre  of  a  rib,  and  entered  the  lungs  close  to  the  heart 
Its  not  having  glanced  on  the  rib  showed  the  advantage 
of  a  hardened  roimd  bullet  over  a  conical  I  had  fired 
with  a  heavy  breach-loading  No.  12,  with  a  charge  of 
five  drachms  of  the  strongest  coarse  powder. 

After  following  up  the  smaller  rhinoceros  for  a  short  dis- 
tance unsuccessfully,  it  became  dusk,  and,  strapping  the 
koodoo  on  a  baggage-mule  of  Mockler  s,  we  started  home- 
warda  Scarcely  had  we  done  so  when  two  lions  began 
roaring  in  some  bushes  close  by.     We  went  up  to  the 


VISIT  TO  KEREN.  123 

place,  but  it  was  too  dusk  to  see  anything.  Scarcely 
had  we  crossed  the  river  when  another  lion  commenced, 
and  we  heard  others  on  our  way  home.  This  was 
during  our  stay  in  the  Anscba,  a  circumstance  of  nightly 
occurrence,  yet  in  the  daytime  we  could  scarcely  ever 
find  the  animals. 

For  two  or  three  days  Jesse  and  I  were  busily  occu- 
pied in  collecting  and  skinning.  We  drew  lots  for  tie 
spoils  of  the  rhinoceros.  Jesse  won  them,  and  with 
much  labour  preserved  the  skeleton,  which  is  now  in  the 
British  Museum,  and  has  been  determined  to  be  that  of 
the  keitloa. 

On  the  16th  Mr.  Munzinger  arrived,  having  ridden 
from  Ailat  in  only  three  days.  On  the  following  mom- 
ing  Mockler  and  I  rode  with  him  to  Keren,  the  principal 
village  of  the  Bogos,  about  sixteen  miles  south-west  of 
Wasentel.  The  road  lies  up  the  Anseba  valley,  the  river 
being  crossed  about  half-way.  Although  we  had  only 
had  occasional  showers,  usually  in  the  afternoon,  heavy 
rain  frequently  fell  near  the  source  of  the  stream  in 
Hamazen,  and  the  water  was  constantly  very  muddy, 
and  the  river  more  or  less  flooded.  x\fter  crossing 
some  hilly  ground,  our  route  lay  up  the  bed  of  the  Dari, 
a  tributary  of  the  Anseba,  now  dry,  and  thence  over 
undulating  ground  to  Keren. 

This  was  the  largest  ^dllage  we  had  seen.  A  consider- 
able proportion  of  the  houses  were  the  usual  beehive-like 
mat  huts,  but  there  were  also  a  number  of  much  larger 
dwelling-places  with  walls  of  wattle,  and  thatched.  The 
thatch,  however,  was  not  very  good.     A  heavy  thunder 


124  PERSONAL  NARRATIVE 

and  hail  storm  came  on  soon  after  we  arrived,  which 
penetrated  through  it,  and  we  were  obliged  to  take  refuge 
in  an  inner  room  covered  with  mats, — a  larger  specimen, 
in  fact,  of  the  usual  Bedouin  beeliive. 

Keren  lies  at  the  foot  of  a  fine  range  called  Zeban, 
consisting  of  highly  granitoid  rock.  All  the  metamor- 
phics  aroimd  are  more  highly  crystalline  than  usual,  and 
the  hUls,  as  usual,  in  consequence  of  the  peculiar  mode 
of  weathering,  appear  to  consist  of  gigantic  rounded 
blocks  piled  upon  each  other. 

We  returned  next  day  to  Bejuk.  Wald  Michele,  the 
ruler  of  Hamazen,  had  threatened  to  march  upon  the  Bogos 
and  "  eat  up  '*  the  country,  because  Rassai,  the  Prince  of 
Tigr^,  had  appointed  a  M.  Ren^,  a  Frenchman  in  his 
service,  to  receive  the  revenue  instead  of  Wald  Michele. 
Mr.  Munzinger  wrote  to  Wald  Michele,  and  induced  him 
to  put  off  his  raid,  and  to  refer  the  matter  in  dispute  to 
Kassai,  as  the  superior  ruler.  This  was  an  admirable 
instance  of  the  influence  which  Mr.  Munzinger  has 
acquired  amongst  these  people  ;  an  influence  which,  it  is 
scarcely  necessary  to  say,  was  of  the  very  greatest  impor- 
tance during  the  progress  of  the  expedition,  and  to  which 
indeed  no  small  portion  of  its  success  was  due.^ 

Although  Wald  Michele  was  perfectly  friendly  to  us, 
we  judged  that  in  the  event  of  disturbances  it  would  be 

*  I  believe  it  is  Dot  too  much  to  say  that  the  march  of  the  army  upon 
Magdala  would  have  been  delayed  by  at  least  a  month  but  for  Mr.  Munzinger's 
aBsifltance.  It  will  scarcely  be  credited  that  the  only  recompense  made  to 
that  gentleman  for  his  services  has  been  the  abolition  of  the  consulate  at 
Massowa,  and  his  consequent  dismissal  from  the*  British  service,  with  a 
formal  letter  of  thanks  from  the  British  Consul  in  Egypt. 


FAUNA  OF  ANSEBA  VALLEY,  125 

wiser  to  keep  a  little  out  of  the  way,  and  we  conse- 
quently made  two  short  marches  down  the  Anseba 
valley,  the  first  to  Waliko,  the  second  to  a  pLace  called 
Hegyargill^,  near  Maregas.  The  country  continued  much 
the  same. 

The  fauna  of  the  valley  is  very  rich  and  varied.  It 
comprises  many  animals  which  appear  to  be  restricted  to 
what  may  be  termed  the  sub-tropical  zone,  and  which 
are  neither  met  with  on  the  coast  nor  in  the  highlands. 
I  obtained  a  magnificent  collection  of  skins,  both  of  birds 
and  mammals.  The  following  will  serve  to  give  a 
general  idea  of  the  principal  animals  found. 

Monkeys  were  not  very  common,  and  the  baboon 
{Cynocejyhalus  hamadryas)  less  than  usual  Cercopi- 
thecus  griseO'viridis  was  occasionally  seen  in  the  trees 
near  the  river.  I  saw  no  bats,  and  wild  cats  were  rare. 
Leopards  were  less  common  than  in  the  rocky  valleys  to 
the  east.  Lions  abounded,  as  already  remarked.  Hyaenas 
(//.  crocuta)  and  jackals  {Cams  mesomelas)  were  the 
commonest  carnivora. 

Amongst  the  pachydermata,  rhinoceroses  have  already 
been  mentioned.  The  white  African  rhinoceros  {R.  simus) 
does  not  occur  in  Abyssinia  or  the  Nile  valley,  so  far  as 
is  known.  Hyraces  are  not  common.  The  wild  hog 
{Phacochcerus)  was  occasionally  met  with,  but  it  is  less 
common  than  near  the  coast.  The  iiiminants  of  the 
Anseba  valley  are  all  antelopes.  The  koodoo  is  found  in 
the  open  valley,  the  klipspringer  on  the  rocky  hills 
around ;  while  the  little  Beni  Israel  {Neotragus  salti- 
arius)    keeps   to   the    bushes.     Another  bush   antelope 


126  PERSONAL  NARRATIVE. 

(perhaps   CephahphiLS   madoqua)   was  also   seen,   but 
not  shot 

Hares  were  scarce ;  those  seen  belonged,  I  think,  to 
the  large  highland  race  (L.  tigrensis).  I  had  ill  success 
with  rats,  and  could  catch  none,  though  I  constantly- 
set  traps  for  them.  Sciurus  anmilatus  was  found  in 
the  river  thickets,  and  Xerus  leuco-umbrimis  in  the 
open  ground. 

Of  raptorial  birds,  the  most  common  were  Gyps  Riip- 
pelK,  the  two  conmion  species  of  Neophron,  Nisus  spJie- 
nurus,  Aquila  rapax,  Milvus  migrans,  and  the  great 
owl  {Bubo  lacteus).  The  magnificent  Hehtarsus  ecau- 
datus  was  not  uncommon,  and  I  shot  one  specimen  of 
Falco  harharus.  Parrots  were  represented  by  Pioniis 
Meyeri  and  PalcBornis  torquata,  barbets  by  Pogono- 
rhynchus  abyssinicus,  Barhatuln  pusilla,  and  Trachy- 
phonus  margaritatus.  One  woodpecker  only  was  seen 
{Picus  nuhiciis),  but  two  species  of  indicator  (/.  minor 
and  /.  Sparmanni)  were  obtained.  Cuckoos  were  more 
numerous,  Centropus  siiperciliosus  being  far  from  rare. 
I  also  shot  Oxylophus  afer,  0.  jacohinus,  Chrysococcyx 
cuprexLS,  and  just  before  leaving,  Cuculus  canorus,  which 
appeared  in  August.  Both  the  plantain-eaters  {Turacus 
leucotis  and  Chizderhis  zonura)  occurred,  and  both  colies 
(C.  leucotis  and  C.  macrourv^).  Neither  Caprimtilgi  nor 
swifts  were  very  common  until  the  beginning  of  August, 
when  the  European  swift  {Cyp>selus  apii^)  made  its 
appearance.  The  three  rollers  abounded  as  at  Kokai ;  of 
kingfishers  Ispidina  2>icta  was  not  rare  in  the  thickets  ; 
Halcyon  senegahnsis  and  H.  semicwrulea  also  occurred. 


FAUNA  OF  ANSEBA  VALLEY.  127 

The  only  bee-eater  found  was  Merops  erythropterus ; 
of  hombills,  Tockics  nasutus  and  T.  erythrorhynchus 
were  common.  T.  Hemprichii  was  only  found  on  the 
hills  around ;  Bucorvus  ahyssinicus  was  frequently  seen. 
I  shot  two  species  of  Irrisor  {I.  erythrorhynchus  and 
/.  aterrimus)^  and  the  common  hoopoe. 

The  passerine  birds  were  of  course  numerous :  it  will 
suffice  to  mention  the  principal.  Swallows  were  repre- 
sented by  Hirundo  puella,  my  new  species  H.  CBthiopicaj 
H.  rusticay  and  Psalidoprogne  pristoptera ;  sunbirds 
by  Nectarinia  affinisy  N.  pulchella,  and,  on  the  hills 
around  only,  N.  cruentata;  flycatchers  by  Tchitrea 
melanogastra  and  Platysteira  pririt ;  shrikes  by  Lanius 
humeralis  (not  common),  Laniarius  cethiopiciis,  L.  gam" 
bensis  in  the  river  thickets  alone,  L.  erythropterus, 
Nilaus  hruhru  and  Dicrurus  divaricatus.  Thrushes 
were  rare  :  Mr.  Jesse  obtained  Turdus  pelios,  and  I  saw 
Cossypha  semirufa ;  Turdus  simensis  did  not  appear  to 
occur  at  this  elevation.  Most  Saxicolince  were  absent  at 
this  season.  The  sylvians  were  represented  by  the  Abys- 
sinian wren,  Oligocercus  micrurus,  and  Camaroptera 
hrevicaudata ;  they]  also  were  not  abundant,  nor  were 
DiymceccB.  Two  species  of  CrateropuSj  C.  Uucocephalus 
and  C  lexicopyghiSy  were  common,  and,  as  in  other 
instances,  exemplified  the  meeting  at  this  elevation  of 
birds  which,  like  the  former,  range  to  the  sea-coast,  and, 
like  the  latter,  to  the  highlands.  Of  the  tits  and  their 
allies.  Par  us  leucopterus,  Zosterops  abyssinica,  and 
Eremomela  grlseo-flava  occurred. 

The  only  crow  was  Corvus  affinis.     The  starling  tribe 


128  PERSONAL  NARRATIVE, 

was  numerously  represented :  Lamprotornis  purpur- 
optera,  Lamprocolins  chrysogastevy  L.  chalybcBus,  Bu- 
phaga  erythrorhyncha,  and  Te^tor  al^^cto,  were  all 
abundant  Tlie  weaver-birds  were  Ilyphantoiviis  galhula 
and  H.  luteola,  both  breeding,  and  Quelea  (ethiopica.  The 
principal  finches  were  Vidua  serenay  Aviodiiia  cantans, 
EstreMa  minima,  E.  phcenicotis,  Pytelia  citerior,  Em- 
heriza  Jlaviventris,  and  Parser  Sivai)isoni.  I  observed 
no  larks,  but  Anthus  sordidus  was  common  on  the 
hill-sides. 

Of  pigeons  and  doves  the  following  occurred  :  Columba 
guinea,  Turtiir  semitorquatus,  Rupp.,  T.  senegalensis, 
Peristera  afra,  and  Treron  ahyssinica.  Guinea-fowls 
and  two  species  of  partridge  {FrancoUnus  Riippelli  and 
F.  gutturalis)  were  common ;  F.  Erkelii  inhabited  the 
hills  around  the  valley. 

Waders  were  scarce.  Scopus  umbretta  was  found 
on  the  river  banks,  with  a  few  sandpipers.  The  only 
plovers  shot  were  (Edicnemus  affinis,  Lohivanellus  sene- 
galensis, both  very  scarce,  and  Sarciophorus  tectum.  The 
only  swimming  bird  was  the  Egyptian  goose,  which  in 
August  began  building  nests  on  the  trees  near  the  river. 

I  obtained  one  land  tortoise  {Cinixys  Belliana),  and 
one  river  species  {Pehmedusa  Gehajie).  Some  scinques 
and  lizards  were  seen,  but  none  captured.  I  found  a 
young  chamjeleon,  apparently  new,  but  allied  to  C.  IcBvi- 
gatus.  The  only  snakes  procured  were  Boddon  lemnis- 
catum  and  Psaiivmophis  sihilans.  Reptiles  altogether 
were  scarce,  and  no  fish  occurred  in  the  stream,  which  is 
dry  for  eight  months  in  the  year. 


LEAVE  THE  ANSEBA  VALLEY.  129 

A  month  passed  rapidly.  The  weather  was  by  no 
means  unpropitious.  Although  this  was  the  height  of 
the  rainy  season,  the  nights  and  mornings  were  invari- 
ably fine.  Eain  usually  commenced  about  two  or  three 
o'clock  in  the  afternoon,  and  lasted  for  one  or  two  hours, 
occasionally,  but  rarely,  until  sunset.  More  frequently 
only  a  shower  fell.  Mockler,  who  devoted  his  attention 
chiefly  to  rhinoceroses,  was  not  very  successful,  but 
before  leaving  the  valley  we  succeeded  in  shooting  a 
second,  rather  smaller  than  the  first.  We  had  abundance 
of  small  game ;  guinea-fowl  and  partridges  were  always 
procurable  when  the  larder  was  empty,  and  Mockler 
killed  several  koodoo ;  all  does,  however,  until  just  as  we 
were  leaving  the  valley,  when  we  shot  a  fine  buck  be- 
tween us.  One  evening  Jesse  and  I,  who  had  returned 
to  camp,  received  a  message  that  Mockler  was  watching 
a  lion  close  by.  We  went  out,  and  found  our  friend 
sitting  on  a  low  hill.  About  300  yards  away,  in  an 
open  piece  of  ground,  lay  an  immense  lion  with  a  short 
mane,  as  usual  in  Abyssinia.  Our  attempt  to  shoot 
him  was,  however,  unsuccessful ;  for,  on  approaching, 
he  slunk  ofi*. 

Our  supplies  gradually  ran  out,  and  the  only  com 
procurable  at  Keren  was  millet ;  even  that  was  very 
scarce.  On  the  8th  August  we  quitted  the  banks  of  the 
Anseba  with  much  regret,  and  crossed  the  Mashalit  ridge 
once  more.  Mr.  Munzinger,  who  had  only  been  able  to 
remain  with  us  part  of  the  time  we  were  in  the  camp, 
had  rejoined  us,  and  also  a  Mr.  Schmidt,  who  was  return- 
ing from  Keren.     Whilst  halting  at  Kokai  on  the  9th,  a 

K 


130  PERSONAL  NARRATIVE. 

lieavy  shower  fell,  and  the  ravine  beside  us  was  suddenly 
filled  with  a  torrent  three  or  four  feet  deep.  In  an  hour 
this  liad  run  down  again. 

On  the  10th  August  we  marched  to  Kelamet.  This 
had  become  much  greener  than  before,  but  still  the 
change  was  great.  On  the  11th,  in  a  most  violent  squaU 
of  wind  and  rain,  several  of  our  tents  were  blown  down, 
including  my  bell  tent  of  course.  A  bell  tent  always  is 
the  first  tent  to  be  blown  down.  Luckily  I  had  secured 
my  collections  in  boxes,  and  only  lost  a  few  butterflies. 

In  consequence  of  these  frequent  showers,  floods  in 
the  Lebka  had  become  of  daily  occurrence,  and  the  gorge 
of  Aualid  Oret  was  impracticable  and  dangerous.  In- 
stead, therefore,  of  returning  by  the  river-bed  as  we 
had  ascended,  it  was  necessary  to  make  a  detour  to  the 
northward,  and  to  go  round  by  Rairo  near  Af  Abed. 
After  the  thorough  wetting  of  the  day  before,  I  was 
occupied  all  the  morning  in  drying  my  tent,  &c.  It  was 
consequently  afternoon  before  I  started  with  Jesse,  all 
our  companions  having  gone  on  earlier.  Leaving  the 
Lebka  valley,  our  route  lay  across  very  open  country, 
over  wide  plains  covered  with  thin  bush  jungle,  and 
scattered  hills.  The  rocks  became  excessively  granitoid, 
and  scattered  tors  of  the  usual  detached  rounded  masses 
were  frequent  We  had  accomplished  the  greater  por- 
tion of  the  march,  which  did  not  exceed  about  fifteen 
miles,  when  a  heavy  shower  induced  us  to  take  shelter 
under  some  rocks  beside  a  dry  stream-bed,  which  we  had 
just  crossed.  Presently  there  was  a  rush  of  water,  and 
the  stream-bed  was  filled  with  a  torrent  about  four  feet 


STOPPED  BY  A  FLOODED  STREAM.  131 

deep,  and  almost,  if  not  quite,  impassable  from  its 
\dolenee.  The  rain  soon  ceased,  and  we  started  again, 
congratulating  ourselves  on  having  been  on  the  right 
side  of  the  stream.  However,  we  had  only  progressed 
about  a  quarter  of  a  mile,  and  the  roaring  of  the  waters 
behind  us  was  still  audible,  when  we  became  aware  of 
the  unpleasant  circumstance  that  there  was  another 
stream  in  front ;  and,  on  reaching  it,  our  disgust  may  be 
conceived  at  finding  that  it  was  broader  and  deeper  than 
the  last,  and  that  crossing  it  was  out  of  the  question. 
We  were  in  the  position  of  Horace's  rustic,  and  had  to 
wait  till  the  river  ran  down ;  however,  rivers  in  Abys- 
sinia do  run  down  in  time.  But  there  was  no  change 
until  nightfidl,  and  as  soon  as  it  was  dark  our  guide 
professed  ignorance  of  the  road,  except  in  daylight.  There 
was  no  help  but  to  bivouac.  We  found  a  small  cave,  the 
l)ottom  of  which  consisted  of  soft  sand ;  a  fire  was  soon 
procured  by  the  help  of  our  guns,  and  a  brace  of  guinea- 
fowl,  which  we  had  fortunately  shot  on  the  road,  cut  in 
[)ieces  and  grilled  on  a  stick,  supplied  us  with  supper. 
The  night,  as  usual,  was  fine,  and  next  morning  the 
stream  had  greatly  subsided,  though  water  was  still 
running  in  it.  We  soon  reached  our  destination,  and 
learned  that  Munzinger  and  Mockler  had  also  lost  their 
way,  and  had  passed  the  night  at  Af  Abed,  in  no  greater 
comfort  than  ourselves. 

At  Rairo  we  found  a  very  large  encampment  of  Be- 
douins belonging  to  the  Habab  tribes,  with  sheep,  oxen, 
and  goats.  They  had  brought  their  flocks  fjx)m  farther 
north,  where  no   rain  had  fallen,   and  where  pasture 

K  2 


132  PERSONAL  NARRATIVE. 

was  consequently  deficient  Lions  had  followed  the 
flocks,  as  usual,  and  there  was  great  roaring  at  night, 
one  very  noisy  fellow  coming  close  to  our  camp,  and 
causing  great  commotion.  He  killed  a  cow,  and  in  the 
morning  the  villagers  discovered  his  lair,  and  told  us  of 
it.  We  went  out,  but  meantime  he  had  retreated,  and 
taken  up  his  abode  on  one  of  the  granite  hills,  behind 
a  huge  boulder.  We  waited  at  the  base  of  the  hill,  and 
the  Bedouins  climbed  over  the  top  from  the  other  side  to 
drive  him  down,  but  he  escaped  along  the  slope.  He 
was  a  splendid  lion,  with  the  usual  short  mane.  Either 
the  same  or  another  returned  to  the  camp  as  soon  as  it 
was  dark,  and  treated  us  to  the  usual  serenade. 

The  country  around  Eairo  is  unusually  granitoid,  and 
the  hills  barren  and  tor-like.  On  the  15th  August  we 
marched  back  into  the  Lebka  valley,  encamping  at  a 
place  called  Wonber  Harattib,  a  little  below  Aualid  Oret. 
I  shot  several  birds  on  the  road  {Gymnoris  canicapillus^ 
Emheriza  Jlaviventrisy  &c.),  and  was  helping  to  skin 
some  of  them,  when  a  camel-man  rushed  in  to  say  that 
some  lions  had  come  down,  and  one  of  them  had  seized 
a  camel.  We  were  all  very  soon  upon  the  spot,  not  a 
quarter  of  a  mile  from  our  camp.  It  wanted  quite  half 
an  hour  to  sunset.  On  a  low  rise  in  front  lay  a  fine  lion, 
a  little  below  him  was  a  second,  whilst  amongst  the 
bushes,  only  sixty  yards  from  us,  was  a  huge  tawny 
mass,  quite  undistinguishable  at  first,  until  a  slight 
movement  of  one  portion  served  to  guide  the  eye.  The 
camel  lay  on  its  side,  and  a  lioness  had  seized  it  by  the 
throat,  and  was  holding  it  down.     The  slight  movement 


SHOOT  A  LI0NSS8,  133 

which  had  enabled  us  to  distingiiish  the  two  animals, 
which  were  exactly  of  the  same  colour,  was  due  to  the 
lioness  tearing  at  the  camel's  throatw  Mockler  fired,  and 
hit  the  lioness ;  I  was  aiming  at  the  moment^  and  as  the 
brute  moved  off  without  a  sound,  my  bullet  went  harm- 
lessly under  her  belly.  Jesse  had  not  distinguished  the 
parts  in  the  confused  ma^  of  tawny  fur,  and,  aiming 
carefully  at  the  lioness,  as  he  thought,  found  subse- 
quently that  his  mark,  which  he  hit,  was  the  unfortunate 
camel's  thigL  The  lioness  ran  up  the  side  of  the  hill, 
under  a  very  heavy  fusUlade,  and  fell  over  dead.  The 
camel  had  several  tooth-marks  in  the  throat,  but  no 
large  blood-vessel  was  injured,  and,  despite  lion  and 
bullet,  she  lived  to  reach  Massowa  some  days  afterwards. 
The  other  lions  ran  off ;  we  followed,  and  fired  several 
shots,  but  they  had  a  long  start,  and  we  could  not  hit 
them.  They  did  not  fail  to  return  in  the  evenings  and 
amuse  themselves  by  roaring  close  to  our  camp  as  usual 

This  was  a  good  instance  of  the  insolence  which  lions 
show.  They  are  certainly  bolder  animals  than  tigers^ 
but  all  the  wild  animals  in  Abyssinia  appeared  to  me 
to  fear  man  less  than  they  do  in  India.  The  only 
exception  is  amongst  the  antelopes,  which  are  at  least  ' 
aB  timid  in  Africa  aa  in  Hindostan. 

On  the  following  day  we  marched  down  the  Lebka  to 
Ain.  The  floods  which  had  descended  had  changed  the 
little  watercourse  greatly,  but  away  from  its  banks  all 
was  parched  and  arid.  There  were  far  fewer  birds ; 
many  appeared  to  have  migrated  to  greener  regions. 

On  the  1 7th  August  we  started  in  the  afternoon  for 


134  PERSONAL  NARRATIVE, 

the  long  marcli  from  Ain  across  the  desert  of  Shob.  Iii- 
st^jad  of  keeping  near  the  hills  we  took  the  direct  road  to 
Massowa,  which  runs  further  to  the  eastward.  It  was  an 
intensely  hot  night,  and  we  had  a  most  dreary  march, 
and  became  one  and  all  intensely  thirsty.  About  two  in 
the  morning,  to  our  surprise,  we  heard  the  rush  of  water, 
and  found  that  a  flood  had  come  dowm  one  of  the 
streams  from  the  hills  called  Mai  Aualid,  and  filled  the 
watercourse  with  liquid  mud  rather  than  water ;  how- 
ever, thick  as  it  was,  it  tasted  delicious  after  our  long  hot 
march.  In  the  morning  Mockler  went  on,  Jesse  and  I 
waited,  intending  to  ride  forward  in  the  afternoon ;  but 
about  one  o'clock  a  sudden  puff  of  wind  carried  away 
my  bell  tent,  and  we  started,  and  rode  about  six  miles 
further  to  the  Amba,  another  watercourse  with  a  few 
pools  of  water  in  the  bed,  near  to  which  stood  Mockler  t:^ 
tent. 

This  is  a  well-known  resort  for  oryx,  and  Mockler  had 
seen  a  herd  in  the  morning.  Jesse  and  I  resolved  to  halt 
there  for  two  or  three  days,  whilst  Mockler  went  on  to 
Massowa,  where  we  expected  a  steamer  would  touch 
about  this  time,  in  which  we  could  all  have  returned  to 
Aden.  The  next  morning  accordingly  both  of  us  went 
out  early  in  separate  directions.  I  went  several  miles, 
and  saw  one  or  two  Soemmering  s  and  Dorcas 
gazelles,  but  no  oryx,  although  then-  tracks  were 
common.  I  had  nearly  returned  to  camp,  when,  ascend- 
ing from  a  small  ravine,  I  at  last  saw  on  the  plain  in 
front  the  animal  I  had  longed  to  catch  sight  of  ever  since 
I  had  been  in  Abyssinia.    Even  at  the  distance  of  hnlf  a 


SUOOT  A  BEISA  ORYX.  135 

mile,  the  cre^am-coloured  hide  and  black  marks  on  the 
legs  and  face  were  sufficient  to  show  what  the  antelope 
was.  I  lay  down  and  watched  her  (it  proved  to  be  a 
female)  through  a  glass.  The  long,  straight,  tapering 
horns,  the  beautifully-formed  limbs,  the  almost  bovine 
faiil,  and  the  singular  symmetry  of  the  Beisa,  as  this  oryx 
is  called,  combine  with  the  unusual  colour  to  render  it 
one  of  the  most  striking  and  elegant  even  of  antelopes. 

The  or}^x  moved  off  with  a  quick,  stei^dy  walk  :  she  had 
seen  me,  and  turned  from  time  to  time  to  look.  Leaving 
my  hoi-se  and  men  for  her  to  watch,  I  stole  up  a  ravine, 
and  at  its  head  lay  l)eliind  some  rocks  till  she  had  crossed 
a  low  rise,  when  I  pushed  on  to  the  top,  and  had  the 
satisfaction  of  seeing  her  walking  slowly  only  150  yards 
away.  I  fired  and  hit  her,  missing  with  my  second 
V>arrel,  liowever,  but  her  leg  was  broken,  and  following  up 
I  hit  lier  again.  A  long  chase  ensued ;  I  was  excited  and 
unsteady  from  running,  and  missed  several  times,  till  at 
last  I  had  but  one  bullet  left,  and  as  wounded  oryx  have 
the  reputation  of  charging,  I  slipped  a  shot  cartridge  into 
tlie  second  barrel  of  my  rifle.  However,  a  Bedouin  cut 
her  off  as  she  was  very  nearly  done,  turning  her  towards 
me,  and  I  had  the  satisfaction  of  rolling  her  over.  I  was 
surprised  to  find  it  was  a  female  :  from  the  splendid 
liorns,  twenty-nine  inches  long,  I  had  certainly  supposed 
it  to  be  a  buck,  but  the  females  of  the  Beisa  oryx  have 
horns  as  long  as  those  of  the  males,  or  longer. 

I  secured  and  preserved  the  skin.  Specimens  of  this 
kind  of  oryx  (0.  Beisa)  are  extremely  scarce  in  museums, 
far  more  so  than  those  of  the  Cape  oryx  (0.  capensis)^ 


136  PERSONAL  NARRATIVE. 

or  of  the  North  African  (0.  leucoryx).  On  the  follow- 
ing afternoon  I  had  a  great  piece  of  good  fortune.  I 
saw  two  of  these  fine  antelopes,  and  made  a  successful 
stalk,  getting  close  to  them,  when  they  perceived  me, 
and  ran  out.  I  dropped  one,  and  the  other  turning 
round  I  shot  it  also,  thus  killing  two  right  and  left.  On 
the  third  day  I  killed  a  fourth.  Jesse,  who  had  worked 
harder  than  I  had,  and  had  gone  over  more  ground, 
singularly  enough  did  not  see  a  single  oryx. 

There  were  a  few  birds  at  Amba  which  I  had  not 
previously  met  with,  amongst  others  Syhria  chierea.  To 
my  surprise  I  saw  the  Abyssinian  roller  also  in  this 
hot  region. 

A  party  of  Egyptian  officers  remained  two  days  at 
Amba  whilst  we  were  there.  They  were  engaged  in 
laying  out  the  line  for  an  electric  telegraph  to  unite 
Massowa  and  Kussala.  Two  of  them  spoke  French,  and 
one  of  them  knew  a  little  English, 

A  few  of  the  migratory  birds  were  beginning  to  return 
from  the  north.  I  shot  a  grey  harrier  {Circtus  cineraceiis) 
on  August  19th,  and  Lanius  isahelUnus  on  the  22d. 
No  SaxicolcB  appeared,  however,  except  Cercomela 
melanura,  which  is  a  resident. 

We  returned  to  Massowa  on  the  23d  August.  No 
steamer  had  arrived  from  Aden,  and  on  the  27th, 
the  Egyptian  steamer  coming  in,  Je^se  left  by  her  for 
Suez  and  England.  Mockler  had  not  been  able  to 
obtain  a  boat  for  Aden,  nor  could  we  secure  one  till 
the  29th,  when  we  shipped  ourselves  and  our  men,  and 
started. 


RETURN  VOYAGE  TO  ADEN.  137 

The  boat  we  had  secured  was  an  open  Arab  dhow  of 
small  size.  There  was  a  quarter-deck,  just  large  enough 
to  lie  upon,  and  a  small  den  below,  in  which  some  of  our 
people  usually  ensconced  themselves.  The  only  covering 
during  the  day  was  a  sail,  but  even  with  this  poor  pro- 
tection the  change  from  the  intense  heat  of  Massowa 
wa^  refreshing. 

Of  course,  for  any  boat  to  start  on  a  cruise  properly 
provided  is  one  of  those  accidents  which  does  not  happen 
in  the  East ;  we  had  scarcely  congratulated  ourselves  on 
having  succeeded  in  only  losing  six  days,  when  we  found 
that  our  craft  had  not  sufficient  supply  of  water  on  board, 
and  that  it  would  be  necessary  to  run  across  to  Dissee 
island,  in  order  to  take  some  in.  This  was  not  the  less 
provoking  that  the  wind  was  fair  to  run  out  past  Dahalac. 
However,  there  was  no  help  :  to  Dissee  we  went,  and 
remained  there  till  next  day  at  noon.  I  atoned  for  the 
delay  by  shooting  a  fine  pair  of  pelicans  (P.  rufescens) 
in  full  plumage,  with  richly  roseate  backs,  and  picking 
up  a  number  of  interesting  shells  and  corals  on  the 
shore. 

At  last,  on  the  30th  August,  we  started  fairly^  for 
Aden.  Our  speed  was  not  great,  for,  if  there  was 
sufficient  wind  to  carry  our  boat  along  at  more  than 
about  six  knots,  our  crew  immediately  shortened  sail,  for 
fear  of  carrying  away  the  mast,  which  was  badly  strained. 
We  had  some  little  difficulty  in  inducing  our  men  to  go 
on  at  night,  and  not  lie  to  altogether,  but  Mockler  per- 
suaded them,  by  the  promise  of  additional  payment,  to 
compromise  the  matter  by  only  shortening  sail.    Luckily 


138  PERSONAL  NARRATIVE. 

the  wind  was  fair  and  light ;  if  our  i)rogrcss  wa.s  not  fast, 
it  was  steady,  and  we  ran  down  to  the  Straits  of  Babel - 
mandel  in  tovo  days  and  a  half,  passing  Perim  Island  on 
the  1st  Septemher.  Thence  to  Aden  our  progress  was 
slower,  tlie  wind  was  unfavourable,  and  we  only  arrived 
in  the  harbour  on  the  morning  of  the  4th.  On  landing 
we  were  most  kindly  received  by  General  Russell,  the 
Resident.  I  WcOS  delayed  for  a  week  waiting  for  a 
steamer  to  Bombay,  which  I  finally  reached  on  the  IStli 
.  September,  after  an  absence  of  nine  months  and  a  half, 
eight  of  which  had  l>een  spent  in  Africa. 

Altogether,  I  have  never  spent  eight  months  more 
pleasantly.  The  country  was  most  interesting,  the 
climate  during  the  greater  part  of  the  time  perfect,  and 
the  fauna  and  geology  had  all  the  attractions  of  novelty. 
The  time  was  quite  insufficient  for  a  thorough  examina- 
tion of  the  country  traversed,  but  still  I  had  reason  to 
congratulate  myself  at  having  been  able  to  accomplish 
what  I  had.  The  scientific  results  of  my  journey  may  be 
thus  briefly  summed  up.  I  had,  I  believe,  succeeded  in 
determining  the  true  succession  of  the  principal  rock 
systems  of  Abyssinia,  and  of  defining  their  characters 
with  much  greater  exactness  than  had  previously  been 
done,  besides  confirming,  upon  a  considerable  amount  of 
additional  fossil  evidence,  the  opinions  of  Me.ssrs.  Ferret 
and  Galinier  as  to  the  age  of  the  only  series  which  has 
yielded  organic  remains.  I  had  collected  about  1,700 
specimens  of  Vertebrata,  representing  350  species,  besides 
a  considerable  number,  about  3,500  specimens,  of  Mol- 
lusca  and  Articulata,  representing  about  500  species.     I 


CONCLUSION.  13J> 

can  only  add  in  conclusion,  that  this  amount  of  success 
^^as  due  to  the  hearty  assistance  which  I  received  from 
the  officers  of  the  army ;  all,  almost  without  exception, 
appeared  to  take  a  genuine  interest  in  scientific  research, 
and  a  desire  to  aid  nie  in  canying  out  the  objects  of 
my  mission  to  Abyssinia.  While  it  is  very  much  to 
be  regretted  that  a  wider  exploration  could  not  bo 
carried  out,  aud  that  no  additional  knowledge  was 
obtained  of  those  parts  of  the  country  to  the  southwards 
which  are  but  little  known,  or  quite  unknown,  to  Euro- 
peans, it  is  a  satisfaction  that  the  campaign  was  not 
allowed,  like  so  many  others,  to  be  entii'ely  useless 
to  science,  and  that  at  least  an  endeavour  was  made  to 
profit  by  the  opportunity  afibrded  of  examining  under 
peculiarly  favourable  conditions  a  most  interesting  region 
of  the  earth's  surface. 


PART  IL 


GEOLOGY. 


S  £ 


GEOLOGY. 


INTRODUCTIOX. 


While  a  considerable  number  of  details  relating  to  the 
geology  of  the  country  ti^aversed  have  been  mentioned 
in  the  preceding  part,  it  is  intended  in  the  following 
pages  to  give  a  general  description  of  the  various  rock 
systems  observed.  These  appear  to  embrace  the  only 
great  fonnation  hitherto  recorded  by  travellers  as  occur- 
ling  in  Abyssinia,  with  the  possible  exception  of  the 
(•oal-beds  said  to  be  found  in  Chelga,  north-west  of  Lake 
Dembea,  of  which,  however,  no  account  hitherto  pub- 
lished appears  to  aflford  any  geological  information,  and 
which  may  possi1)ly  prove  to  be  part  of  the  same  system 
as  the  Adigrat  sandstone. 

Before  passing  to  these  descriptions,  a  few  words  on 
the  lal)oui's  of  previous  observers  are  necessary.  The 
geology  of  Al)yssinia  has  met  with  far  less  attention 
than  its  zoolog}",  and,  with  the  exception  of  Messrs. 
Ferret  and  Galinier,  no  travellers  have  left  any  very 
extensive  record  of  their  observations.   Scattered  remarks 


144  OEOLOOT. 

on  the  rocks  are  distributed  through  many  works,  from 
those  of  Bruce  and  Salt  to  the  recent  narratives  of 
Markham  and  Von  Heuglin;  but  the  only  important 
contributions  to  the  geology  of  the  country  with  which 
I  have  been  able  to  meet  are  the  following : — 

In  1834  RUppell  published  a  "  Sketch  of  the  Geological 
Formations  of  Abyssinia*'  (**Skizze  der  geologischen 
Formation- Abyssiniens "),  in  the  Museum  Senkenher- 
gianum^  voL  L  p.  286.  It  is  merely  a  sketch,  as  its 
name  implies,  but  a  good  one  nevertheless,  and  contains 
much  information  on  the  diflFerent  rocks  found  on  the 
shores  of  the  Red  Sea,  and  in  Agam^,  Adowa,  Temben, 
Simen,  and  around  the  Lake  Tzana  or  Dembea.  All 
the  Simen  hills  and  the  whole  country  around  Lake 
Dembea  consist  of  volcanic  rocks.  I  must  differ,  how- 
ever, from  Dr.  Ruppell  on  one  point.  He  speaks  of  the 
massive  hills  of  Senaf^  as  of  hardened  marl  (Kalkmergel). 
As  already  stated,  they  appeared  to  me  to  be  of  a  kind 
of  claystone  passing  into  trachyte. 

A  letter  from  M.  d'Abbadie,  published  in  the  Bulletin 
de  la  Soc.  Giologique  for  1839,  vol  x.  p.  121,  gives, 
in  very  few  words,  an  excellent  description  of  the  coast 
near  Massowa  (which  he  considers  an  alluvial  formation 
daily  increased  by  the  effects  of  torrents),  and  of  some 
geological  phenomena  observed  on  the  route  between 
Massowa  and  Gondar.  The  high  temperature  of  the 
springs  in  Samhar  is  noticed,  and,  on  the  highlands,  the 
remarkable  masses  of  sandstone  with  perpendicular  faces. 
In  M.  d'Abbadie's  opinion,  the  form  of  the  mountains  of 
Tigrd  is  not  due  to  the  effect  of  tropical  rain. 


INTRODUCTION.  146 

A  much  longer  "  Rapport  geologique  et  mindralogique 
sur  la  Province  de  Tigre,"  by  M.  Vignaud,  published  in 
the  same  Bulletin  for  1843,  vol.  xiv.  p.  492,  consists 
mainly  of  an  account  of  the  country  around  Adowa.  The 
volcanic  rocks  in  the  neighbourhood  of  the  Red  Sea  are 
mentioned,  and  the  destruction  of  ancient  Adulis  attri- 
buted to  a  **  general  movement "  connected  with  them, — 
a  view  which  appears  to  have  found  favour  with  several 
travellers,  but  for  which  I  am  unable  to  find  any 
authority.  Most  of  the  volcanic  rocks  on  the  shores  of 
the  Red  Sea  are  certainly  far  more  ancient  than  Adulis, 
and  the  destruction  of  the  town  is  more  probably  due  to 
political  than  to  volcanic  causes.  M.  Vignaud^s  petro- 
logical  nomenclature  is  so  peculiar,  that  his  account  is 
not  easUy  understood.  He  gives  a  minute  description  of 
the  rocks  met  with  between  Adowa  and  the  Mareb,  and 
appears  to  regard  all,  including  the  volcanic  formations, 
as  of  transition  age.  Unlike  M.  d'Abbadie,  be  considers 
that  the  violent  rains,  by  removing  the  soluble  portion  of 
the  rocks,  such  as  alumina  and  lime,  have  hollowed  out 
the  ravines. 

In  the  Comptes  Rend  us  for  1841,  vol.  xii.  p.  732,  is 
an  abstract  of  a  memoir  by  M.  Rocher  d'Hdricourt,  on  his 
first  journey  from  Tajurra  to  Shoa.  He  especially  refers  to 
the  presence  of  volcanic  rocks,  and  of  hot  springs.  A 
brief  account  of  the  second  journey  of  the  same  writer 
appears  in  the  Coinptes  Rendus  for  1845,  vol.  xxi.  p.  883. 
It  contains  nothing  of  geological  importance  ;  but  in  the 
Bulletin  de  la  SociSte  Geologique  de  France  for  1846, 
p.  541,  the  same  author  gives  a  brief  but  still  much  more 


146  QEOLOQY. 

complete  account  of  his  geological  observations  made  in 
Egypt,  the  Red  Sea,  the  Gulf  of  Aden,  the  Adel  country, 
and  in  Shoa.  He  describes  the  prevalence  of  volcanic 
rocks,  and  of  recent  cones  in  the  southern  portion  of 
the  Red  Sea ;  also  at  Tajurra,  and  throughout  the  Adel 
country.  He  found  the  basement  rocks  of  Shoa  to  con- 
sist of  granite  and  other  associated  crystalline  rocks, 
upon  which  rest  cones  (?)  of  trachyte  and  basalt,  some- 
times of  great  height  and  vast  extent.  He  especially 
describes  the  hill  above  Angobar  (or  Ankobar),  on  the 
slope  of  which  the  town  is  built,  as  a  crater  of  elevation, 
the  mass  formed  of  trachyte,  while  basalt  covers  the 
sides ;  Angolala,  the  second  capital,  is  built  on  a  hutte 
de  souUvement^  of  which  the  principal  element  is  basalt ; 
whilst  at  Debrabrame,  between  the  two,  zirconiferous 
syenite  is  found.  He  also  describes  a  great  ravine, 
fourteen  leagues  north-west  of  Angolala,  1,254  metres 
deep,  and  only  800  broad.  The  sides  are  of  volcanic 
tufa,  resting  on  primitive  rocks. 

It  is  probable  that  the  traps  of  Shoa  are  precisely 
similar  to  those  to  be  presently  described,  as  occurring 
near  Antalo,  Ashangi,  Magdala,  &c.,  and  that  the 
trachyte  of  Ankobar  really  is  superior  to  the  basalt,  and 
rests  upon  it,  the  rocks  being  probably  bedded,  as  they 
are  further  north. 

Messrs.  Ferret  and  Galinier,  two  staflF-officers  of  the 
French  army,  were  in  1839  entrusted  with  a  commission 
from  their  Government  to  explore  portions  of  Abyssinia, 
and  they  remained  in  that  country  nearly  two  years, 
from  October  1840   till  August  1842.     A  short  notice 


INTRODUCTION.  147 

of  their  geological  observations  appeared  in  the  Comptes 
Reiidits  for  1844,  vol.  xix.  p.  881 ;  but  a  much  more 
extended  account  was  given  in  the  third  volume  of  their 
"Voyage  en  Abyssinie,"  published  in  1847.  The  first 
seventy-six  pages  of  this  volume,  which  is  filled  with 
the  scientific  results  of  the  journey,  are  devoted  to  the 
geology.  Numerous  remarks  on  the  same  subject  are 
scattered  throughout  the  first  two  volumes,  containing 
the  narrative  of  the  personal  adventures  of  the  two 
travellers;  and  the  accompanying  atlas  comprises  a 
geological  map  of  Tigr6  and  Simen,  and  several  geo- 
logical sections. 

The  work  of  Messrs.  Ferret  and  Galinier  is  by  far  the 
most  important  contribution  to  the  geology  of  Abyssinia 
which  has  hitherto  appeared,  and  the  greatest  credit  is 
due  to  those  gentlemen,  not  only  for  the  great  energy 
and  research  displayed  by  them,  but  also  for  the  ad- 
mirably clear  description  of  the  observations  which  they 
have  furnished.  That  they  had  but  little  previous  expe- 
I'ience  of  field  geology  is  probable,  and  I  am  obliged  to 
differ  from  some  of  the  conclusions  at  which  they  have 
arrived  ;  but  the  value  of  their  observMions  is  in  no  way 
impaired. 

Messrs.  Ferret  and  Galinier  divide  the  rocks  observed 
by  them  into  the  following  systems : — 


'^  1.  Terrains  piimalres .     .     T.  gneissique    . 


"  granite 
gneiss 
micaschiste 
protogine 
talcsdiiste 


L  I 


148 


GEOLOOT. 


2.  Terrains  intenn^diaires 


talcschiste 
T,  inf^rieor  ou  phylladiqne    \  phyllade 

Bj^nite 


3.  Terrains  secondoires  . 


T.  carbonique 


diorite 
amphibolitey  &c. 


T.  oolitique 

T.  s^dimentaires  . 


4.  Terrains  tertiaires .    . 


6.  Terrains  ipodemes.*' 


.    .     .       gr^,  &c. 

Roches  d'origine  ign^e  qui  \  trachytes 

•    nent    aux    teiv  j  ^^^ 

iaires  .    .    .    .  i 


appartiennent 
rains  tertiaires 


The  oolitic  rocks  of  this  table  are  the  limestones  near 
Chelikot  and  Antalo,  and  the  sedimentary  tertiary  beds 
the  sandstones  of  Tigr^  described  below  as  the  Adigrat 
sandstones,  to  which  I  assign  a  much  earlier  geological 
age — ^but  they  are  not  fossiliferous. 

Messrs.  Ferret  and  Galinier  also  consider  that  there  is 
evidence  of  the  existence  of  rocks  of  carboniferous,  tri- 
assic,  and  cretaceous  age  in  Abyssinia ;  the  existence  of 
the  first  being  proved  by  the  presence  of  coal,  and  that 
of  triassic  rocks  by  the  occurrence  of  rook-salt  and 
sulphur  in  the  country  between  the  highlands  and  the 
Eed  Sea.  The  indications  of  beds  of  cretaceous  age  are 
considered  as  very^vague ;  but  the  travellers  are  inclined 
to  atlxibute  certain  rocks  in  Shoa,  described  by  M. 
Kocher  d'H^ricourt,  to  this  series.  The  evidence  in  these 
three  cases  would  not  be  deemed  sufficient  in  the  present 
day  to  prove  the  presence  of  the  formations  supposed 
to  occur. 

I  have  classed  together  the  whole  of  the  crystalline 
and  sub-crystalline  rocks,  attributed  by  Messrs.  Ferret 
and  Galinier  to  primary,  intermediate,  and  carboniferous, 


INTRODUCTION.  149 

under  one  series,  the  metamorphie  ;  not  because  I  think  it 
at  all  improbable  that  they  may  be  divided,  but  because 
I  have  not  myself  seen  any  proofe  of  distinction.  The 
oolitic  beds  of  Messrs.  Ferret  and  Galinier  are  my  Antalo 
limestones,  and  I  have  entirely  confirmed  the  opinions  of 
those  explorers.  Their  tertiary  sedimentary  rocks,  as 
already  mentioned,  I  consider  pre-jurassic,  and  the  ter- 
tiary igneous  rocks  of  their  classification  appear  to  me, 
in  part  at  least,  probably  of  somewhat  older  date ;  but 
their  epoch  is  uncertain. 

I  do  not  think  it  necessary  to  do  more  than  refer  to 
the  works  of  Lefebvre,^  or  to  Dr.  Beke's  valuable  papers 
in  the  Royal  Geographical  Society's  Journal^  because, 
although  numerous  geological  notices  occur  in  them,  they 
add  but  little  to  what  was  previously  known  of  Abys- 
sinian geology.  The  same  remark  applies  to  the  works 
of  other  travellers  in  an  even  greater  degree. 

Quite  recently,  in  the  Zeitschrift  der  Ge^ellschaft  fiir 
Erdkuiiide  zu  Berlin,  1869,  p.  347,  a  paper  has  appeared, 
entitled  "  Geological  Sketch  of  the  Neighbourhood  of 
Axum  and  Adowa,  in  Tigre,"  by  Dr.  A.  Sadebeck.  It 
consists  principally  of  notes  on  a  collection  of  rock  speci- 
mens, sent  by  Dr.  W.  Schimpcr  to  accompany  a  map  and 
sections.  The  rock  specimens  comprised  granite  of  two 
forms,  various  crystalline  rocks,  claystone,  claystone- 
breccia  and  conglomerate,  and  sandstone.    The  claystones 

^  Voyage  en  Abyssinie  ezdcut^  pendant  les  Annies  1839-43.  Paris, 
1845. 

*  Vol.  X.  p.  583  ;  xii.  p.  84 ;  xiiL  254 ;  xiv.  p.  1.  As  mentioned  subsequently, 
Dr.  Beke  has  most  obligingly  allowed  me  to  examine  an  interesting  series  of 
rock  specimens  collected  during  his  travels 


150  GEOLOGY. 

are  doubtless  a  variety  of  the  trachytic  rocks ;  but  it  is 
curious  that,  even  when  studied  in  the  cabinet,  the 
author  of  the  paper  has  evidently  been  in  great  doubt  as 
to  whether  they  were  of  sedimentary  or  volcanic  origin, 
and  has  preferred  the  former  opinion.  Having  merely 
rock  specimens,  Dr.  Sadebeck  has  classed  together,  under 
the  general  head  ^of  eruptive  rocks,  granites,  porphyries, 
and  basalts — ^a  view  not  in  agreement  with  the  observa- 
tions made  a  little  farther  to  the  east ;  for  it  is  evident 
that  all  the  granites  observed  belong  to  a  far  more 
ancient  series  of  rock  formations  than  the  basalts. 


SECTION   I. 

PHYSICAL  QEOORAPUr  OF  ABYSSINIA,  AND  ITS  RELATIONS 
TO  THE  GEOLOGY y  WITH  REMARKS  ON  DENUDATION  AND 
ON  LAKE  ASHANGL 

The  physical  geography  of  Northern  Abyssinia  is  well 
known,  and  there  are  few  parts  of  the  world  neither 
inhabited  nor  frequently  visited  by  Europeans  of  which 
so  much  had  been  ascertained  previous  to  the  expedition. 
Beyond  a  few  corrections  of  latitudes  and  longitudes 
very  little  has  been  added  except  in  matters  of  detail. 

Abyssinia  Proper  consists  of  a  mass  of  mountainous 
country,  intervening  between  the  basin  of  the  Nile  and 
the  east  coast  of  Africa.  It  rises  on  the  east  rather 
abruptly  from  the  low  country  bordering  the  Red  Sea 
and  the  Indian  Ocean,  and  slopes  away  more  gradually 
to  the  westward,  the  Nile  tributaries  cutting  out  deep 
valleys.  The  average  height  of  the  dividing  range  be- 
tween the  streams  flowing  to  the  ea^t  and  to  the  west 
is  about  8,000  feet,  rising  to  10,000  or  11,000  to  the 
south  and  sinking  to  the  north.  Some  peaks  in  Simen 
are  said  to  reach  15,000  feet,  and  many  in  various  parts 
of  the  country  are  12,000  and  13,000,  while  to  the  south- 
ward many  plateaux  of  considerable  extent  arfe  more  than 
10,000  feet  above  the  sea. 


152  GEOLOGY. 

The  plateau  contracts  greatly  towards  the  Dorth.  The 
last  of  the  great  affluents  of  the  Nile  in  this  direction  is 
the  Mareb,  which  rises  in  Hamazen,  almost  due  west  of 
Massowa.  The  streams  running  from  the  northern  side 
of  Hamazen,  the  principal  of  which  are  the  Anseba  and 
the  Barka,  unite  and  flow  into  the  Ked  Sea  south  of 
Suakim.  Although  a  few  isolated  plateaux  of  7,000  to 
8,000  feet,  such  as  those  of  Mensa  and  Marea,  exist  in 
the  area  drained  by  these  rivers,  the  elevated  country 
is  of  smaD  extent,  and  the  greater  portion  of  the 
surface  does  not  rise  above  4,000  to  5,000  feet  above 
the  sea. 

The  fundamental  rocks  of  all  Tigr^,  and  probably  of 
all  Abyssinia,  are  metamorphic.  They  compose  the  mass 
of  the  table-land,  and  while  they  occupy  no  inconsider- 
able proportion  of  its  surface,  they  are  exposed,  in  Tigr^ 
at  least,  in  every  deep  valley. 

As  a  general  rule,  it  may  be  asserted  that,  in  the  neigh- 
bourhood of  the  route  followed  by  the  British  army,  so 
much  of  the  country  as  exceeds  a  height  of  about  8,000 
feet  above  the  sea  consists  of  bedded  traps.  There  are 
a  few  exceptions,  such  as  the  sandstone  range  of  Sowera, 
near  Senaf^,  but  they  are  neither  numerous  nor  impor- 
tant The  traps  of  Ashangi  also  descend  to  a  much 
lower  level  than  8,000  feet  above  the  sea.  The  greater 
portion  of  Central  and  Southern  Abyssinia,  as  far  south  as 
Shoa  at  least,  is  known  to  consist  of  the  same  rock.^ 

Between  the  traps  and  the  metamorphics  a  series  of 
sandstones  and  limestones  intervenes,  one  group  of  the 

^  It  is  shown  by  M.  Rocher  d'H^ricourt  to  be  the  case  in  Shoa. 


PHYSICAL  GEOGRAFHT  OF  ABYSSINIJ.  153 

fonner  underlying   the  latter.     The  limestone  aJone  is 
fossiliferous,  and  is  of  Jurassic  age. 

The  country  below  the  mountains  along  the  Ked  Sea, 
inhabited  by  the  Danakil,  Somali,  and  Galla  tribes,  is 
but  little  known,  and  no  opportunity  was  aflforded  to  me 
of  penetrating  into  it  during  the  progress  of  the  expedi- 
tion, the  only  exploration  made,  that  of  Colonels  Phajnre 
and  Mcrewether,  in  the  direction  of  the  Alelbad  Salt 
Lake,  having  been  accomplished  before  my  arrival  at 
Zulla.  The  region  appears  to  consist  of  transverse 
(E.  and  W.)  chains  of  hills,  separating  desert  tracts, 
many  of  which  are  below  the  sea-level,  and  some  of 
which  may  formerly  have  been  parts  of  the  sea  subse- 
quently cut  oflf  and  dried  up.  Water  is  scarce,  the  rain- 
fall being  scanty,  and  all  brought  down  from  the  hills  by 
streams  appears  to  be  evaporated.  Along  the  coast  of  the 
Ked  Sea  there  is  a  belt  of  volcanic  rocks  of  very  late 
geological  age,  the  elevation  of  which  may  have  cut  off 
such  tracts  as  the  salt  plain  west  of  the  Amphila  and 
Lake  Assal  west  of  Tajun-a,  both  of  which  are  below  the 
sea-level. 

This  belt  of  volcanic  rocks  is  extremely  narrow  in  the 
neighbourhood  of  Zulla  and  Massowa.  A  few  observa- 
tions of  interest  were,  however,  made  upon  it,  which  will 
be  found  hereafter.  Before  concluding  this  brief  sketch, 
there  are  two  points  connected  with  the  physical  geo- 
graphy of  Abyssinia  of  which  some  notice  should  be 
taken.  These  are  the  evidence  of  denudation  afforded  by 
its  surface  and  the  conditions  under  which  the  only  lake 
examined  occurs.  Both  are  subjects  of  considerable  interest. 


1 54  GEOLOGY. 

The  surface  of  the  AbyssiDian  plateau,  as  has  already 
been  stated,  has  been  deeply  scored  by  the  valleys  in 
which  the  rivers  flow.  Many  of  those  in  Northern 
Abyssinia  are  remarkable  enough :  the  Takkazzy^  and 
Mareb  flow  in  ravines  at  least  3,000  feet  in  depth.  But 
it  is  in  the  area  of  bedded  traps  that  these  river  valleys 
assume  their  most  peculiar  characters.  The  gorges  of 
the  Jitta  and  Bashilo,  close  to  Magdala,  impressed  every 
one  who  saw  them  by  their  great  depth  and  the  excessive 
steepness  of  their  sides,  the  breadth  being  singularly  small 
in  comparison.  Similar  deep  gorges  are  known  to  be 
found  further  south.^ 

No  such  gorges  can  possibly  result  from  marine  denu- 
dation. The  fiords  of  Norway,  which  are  far  broader  in 
proportion  to  their  depth,  are  glacier  valleys,  and  certainly 
not  due  to  the  action  of  the  sea.  There  is  not  the 
slightest  evidence  of  any  physical  disturbance  which 
could  have  produced  such  rents  in  the  earth's  surface ; 
the  strata  which  cap  the  scarps  at  each  side  of  the  ravines 
are  perfectly  horizontal.  There  has  aa  certainly  been  no 
faulting,  at  all  events  in  the  case,  of  the  Jitta ;  bed  corre- 
sponds  to  bed  on  the  opposite  sides  of  the  gorge  in  the 
most  perfect  manner.  There  cannot  be  a  question  but 
that  these  enormous  hollows  are  simply  channels  cut  by 
the  streams  which   run  in   them.     The  lapse   of  time 

^  One  18  described  by  M.  Rocher  d'H^riconrt  as  traye^sing  part  of  Shoa,  at 
a  distance  of  fourteen  leagues  north-west  of  Angolala,  one  of  the  principal 
towns.  This  ravine  is  said  to  be  1,254  metres  (about  4,000  feet)  in  depth, 
and  only  700  to  800  metres  (2,500  feet)  in  breadth  !  It  is  to  be  presumed 
that  the  measurements  were  taken  at  a  spot  where  the  width  was  exception- 
ally small. 


PHYSICAL  GEOGRAPHY  OF  ABYSSINIA.  155 

necessary  to  have  produced  such  an  eflfect  must  have 
been  very  great. 

If  the  action  of  such  small  streams  as  the  Jitta  and 
Bashilo  has  sufficed  to  sweep  away  the  contents  of 
ravines  3,000  or  4,000  feet  in  depth,  what  may  not  have 
been  the  eflfect  of  rivers  like  the  Takkazzy^  and  Mareb  ? 
How  much  of  the  Abyssinian  highlands  has  been  removed 
by  these  great  torrents,  and  spread  as  an  alluvial  deposit 
over  the  basin  of  the  Nile  ? 

There  is  every  probability  that  the  trap  hills  of  Adi- 
grat,  described  in  a  subsequent  page,  are  a  mere  outlier 
of  the  great  mass  of  bedded  trach}"tes  and  dolerites  of 
Lasta,  Amhara,  and  Shoa.  It  appears  probable,  from 
the  accounts  given  by  travellers,  that  the  high  ranges  of 
Samyen  or  Simen,  south-east  of  Takkaz^y^,  are  another 
similar  outlier.  If  this  be  so,  then  over  a  portion,  pro- 
bably over  the  whole,  of  Northern  Abyssinia,  there 
existed  at  least  4,000  feet  of  bedded  traps,  of  which  now 
only  a  few  vestiges  remain. 

It  is  worthy  of  repetition  that  throughout  this  great 
denuded  area,  so  far  as  it  was  possible  to  examine  it, 
there  is  not  a  trace  of  marine  denudation.  The  prevail- 
ing features  of  the  country  are  deep  ravines  cut  by 
rivers,  and  terraced  hill-sides,  moulded  by  the  sub-aerial 
disintegration  of  the  rocks  of  which  they  are  composed. 
On  all  the  slopes  there  are  unequivocal  marks  of  the 
unequal  action  of  surface  weathering  upon  rocks  of  dif- 
ferent chemical  constitution.  This  is  especially  seen  upon 
the  traps. 

How  far  the  great  scarp  of  the  Abyssinian  plateau 


156  GEOLOGY, 

towards  the  Red  Sea  is  due  to  marine  action  is  a  very- 
difficult  question.  If  the  sea  ever  exerted  any  great 
influence  on  the  denudation  of  the  plains  at  the  base  of 
the  range,  it  must  have  been  at  a  comparatively  remote 
period ;  for  so  far  as  they  were  examined,  the  hills  pre- 
sented a  very  different  aspect  from  that  of  a  sea  cliff. 
Still  it  is  highly  probable  that,  before  the  commencement 
of  the  volcanic  outbursts  which  have  left  their  traces 
along  the  whole  southern  portion  of  its  shores,  the  sea 
extended  farther  west,  and  very  probably  it  reached  the 
foot  of  the  hills  in  places.  But  tropical  seas,  and  espe- 
cially calm  laud-locked  basins,  like  the  Red  Sea,  are 
trifling  agents  of  denudation  when  compared  with  such 
oceans  as  the  Atlantic.  Independently  of  the  local 
causes  which  determine  the  size  of  the  breakers,  most 
tropical  coasts  are  more  or  less  protected  either  by  coral 
reefe,  where  no  sediment  is  brought  down  by  rivers,  or 
by  the  deposition  of  that  sediment  itself,  which,  owing 
to  the  more  violent  rainfall  (a  far  more  important  item 
than  the  absolute  amount,  although  the  latter  is  also 
usually  excessive  in  the  tropics),  and  to  the  more  rapid 
disintegration  of  rock,  is  far  larger  in  quantity  than  in 
temperate  regions.^ 

^  I  believe  that  the  comparative  importance  of  fresh-water  and  marine 
denudation  is  to  this  day  misunderstood  by  many  of  the  best  geologists  of 
Western  Europe,  and  is  only  beginning  to  be  appreciated  by  any  of  them,  in 
consequence  of  the  exceptional  conditions  there  occurring.  The  whole  of  the 
circumstances  attending  the  rainfall  in  tropical  countries  are  far  more  favour- 
able to  denudation  than  in  the  temperate  zone.  Compare  more  especially 
the  case  of  the  British  islands  and  the  neighbouring  portions  of  the  Con- 
tinent— ^the  part  of  the  world,  that  is,  in  which  by  far  the  greatest  number  of 
careful  and  continuous  observations  upon  such  phenomena  as  denudation  have 


PHYSICAL  QEOGRAPHT  OF  ABYSSINIA.  157 

Of  the  lakes  of  Abyssinia,  only  one  was  reached  by 
the  expedition;  and  this,  Lake  Ashangi,  unfortunately 

been  made — ^with  that  of  any  well-known  tract  in  the  tropics,  such  as  British 
India.  I  select  this  for  comparison,  because  I  am  best  acquainted  with  the 
fiicts,  and  do  not  need  to  take  them  at  second-hand.  In  the  British  islands 
the  average  rainfall  is  about  twenty-four  inches  distributed  oyer  the  greater 
portion  of  the  year.  In  India  it  averages  over  the  whole  country  about  fifty 
inches,  by  far  the  greater  portion  of  which  fsdls  in  three  months.  The  showers 
are  far  heavier,  and  far  more  effective  in  sweeping  soil,  sand,  and  pebbles  from 
the  surface  of  the  country  into  the  streams  ;  and  floods  in  {he  latter  are  of 
annual  occurrence,  instead  of  only  happening  at  rare  intervals.  The  effect  of 
a  river  in  full  flood  in  sweeping  detritus  down  into  the  sea  compared  with  the 
usual  denudating  action,  is  as  the  comparison  of  the  effect  produced  by  the 
breakers  of  the  ocean  in  a  storm  to  those  of  an  inland  sea  on  an  ordinarily 
fair  day.  In  flood,  a  river  is  liquid  mud  rather  than  water.  Buty  in  addition 
to  the  great  floods,  minor  floods  in  tropical  streams  are  matters  frequently  of 
daily  occurrence  during  the  rainy  season.  That  precisely  the  same  pheno- 
mena occur  in  Abyssinia  as  in  India  I  had  ample  opportunities  of  seeing  in 
July  and  August ;  indeed,  they  are  notoriously  the  rule  in  all  tropical 
countries,  except  the  few  rainless  regions. 

But  this  is  only  one  element  in  the  comparison.  It  must  be  borne  in  mind 
that  fresh- water  denudation  is  distinctly  antagonistic  to  marine ;  and  where 
the  former  is  large,  the  quantity  of  detritus  carried  down  to  the  sea  by 
rivers  actually  protects  the  rocks  of  the  coast  from  destruction.  The  coasts  of 
Abyssinia  are  so  little  exposed  and  so  much  defended  by  coral  islands  that  they 
could  not  under  any  circumstances  afford  a  fair  opportunity  for  judging,  even 
if  the  drainage  of  the  country  were  not  in  the  other  direction.  The  coasts  of 
the  peninsula  of  Hindustan,  however,  are  an  excellent  case  in  point.  Along 
the  whole  eastern  side  of  the  peninsula,  from  the  head  of  the  Bay  of  Bengal 
to  isii  south  of  Tanjore,  at  all  events,  the  whole  coast  is  fringed  by  alluvial 
deposits,  except  in  one  spot,  close  to  Vizagapatam,  where  alone  the  sea  washes 
away  rock.  Nearly  all  the  great  rivers  of  the  peninsula  pour  their  detritus 
into  the  sea  on  the  east  coast ;  but  even  on  the  west  coast,  where  fewer  and 
smaller  streams,  for  the  most  part,  enter  the  sea,  there  is  throughout  a  very 
large  extent  of  shore  a  similar  alluvial  belt.  Further  north,  in  Guzerat, 
Kattiawar,  Cutch,  and  Sinde,  the  coast  is  in  most  places  formed  of  alluvial 
deposits  derived  from  the  large  rivers  Indus,  Nerbudda,  Taptee,  Mhyr,  &a  ; 
and  throughout  the  whole  immense  shore  line  of  British  India,  exclusive  of 
the  Burmese  provinces,  the  only  places  where  any  marine  denudation  takes 
place  are  between  Bulsar,  south  of  Surat,  and  Mangalore,  and  at  a  few 
isolated  points  further  south  on  the  west  coast,  and  at  Yiaagapatam  on  the 


158  GEOLOGY, 

is  one  of  the  smallest,  although  very  interesting.^  It  is 
four  miles  long  by  three  broad,  and  was  found  by  Major 
Goodfellow  and  a  party  of  engineer  officers,  who  sounded 
it,  to  be  103  feet  deep.  As  soundings  were  only  carried 
on  over  a  small  portion  of  the  lake,  this  may  perhaps 
not  be  the  maximum  depth.  The  spot  at  which  it 
occurred  was  not  far  from  the  south-west  shore,  and 
the  bottom  was  found  to  rise  in  all  directions  around. 
Its  chief  peculiarity  is  the  absence  of  any  perceptible 
outlet  whatever.  To  the  west  are  high  hills,  to  the  east 
a  lower  range.  The  lowest  portions  of  the  rim  sur- 
rounding  the  lake  are  to  the  north-east  and  to  the  south. 
The  latter,  evidently  the  lowest  of  all,  are  about  150  feet 
above  the  lake. 

It  appears  almost  certain  that  the  lake  has  a  sub- 
east.  If  over  British  India  the  effects  of  marine  to  those  of  fresh-water 
denudation  in  removing  the  rocks  of  the  oountiy  be  estimated  at  1  to  100,  I 
believe  that  the  results  of  marine  action  wiU  be  greatly  overstated. 

Some  years  ago,  before  the  question  had  assumed  the  present  phase,  before 
even  Colonel  Greenwood's  pamphlet  had  appeared,  I  remember  being  struck 
by  the  absence  of  all  signs  of  marine  action  and  by  the  unmistakeable  evi- 
dence of  immense  firesh-water  denudation  in  the  Himalayas  of  Sikkim,  where 
ravines  firom  6,000  to  16,000  feet  in  depth  are  evidently  the  excavations  of 
the  rivers  running  in  them :  so  I  am  no  new  convert  to  a  belief  in  the  complete 
efficacy  of  rain  and  rivers  to  produce  gigantic  effects.  But  after  seeing,  both 
in  India  and  Abyssinia,  what  the  effects  of  these  agents  are  in  tropical  coun- 
tries, I  do  not  feel  surprised  that  their  powers  should  be  recognised  with 
difficulty  in  regions  where  their  effects  are  comparatively  so  dwarfed  as  in  the 
British  isles,  while  the  power  of  marine  denudation  is  at  its  mATimnm  from 
the  enormous  coast  line  exposed  and  the  small  amount  of  detritus  furnished 
for  its  protection  by  rivers  of  small  length,  and  in  which  floods  are  of  excep- 
tional occurrence. 

1  One  of  the  principal  geological  questions  which  I  had  hoped  to  investigate 
in  Abyssinia  is  the  mode  of  origin  of  one  of  the  large  African  lakes,  such  as 
Dembea.    Unfortunately,  I  was  unable  to  reach  so  fax. 


PHYSICAL  QEOGRAFHT  OF  ABYSSINIA.  159 

terranean  outlet,  for  two  reasons.  The  first  is,  that 
although  the  army  was  on  its  banks  in  April,  at  the 
close  of  the  driest  season  in  Abyssinia,  there  was  no 
appearance  of  the  lake  having  been  at  any  period  of  the 
year  more  than  a  few  feet  higher  than  it  then  waa 
Situated  as  it  is,  almost  at  the  edge  of  the  steep  descent 
to  the  eastward,  with  high  hills  to  the  west  and  north- 
west, the  rainfall  must  be  very  heavy  ;  and  it  is  difficult 
to  conceive  that  in  the  months  of  July,  August,  and 
September  it  can  be  less  than  eighty  inches,  which  is 
sufficient  to  increase  the  waters  of  the  lake  by  thirty  or 
forty  feet.^  So  far  as  it  was  possible  to  judge,  there  was 
no  evidence  of  the  water  having  been  more  than  a  third 
of  that  amount  above  the  lowest  leveL  It  is  extremely 
improbable  that  the  rainfall  could  be  evaporated  within 
the  year  from  the  surface  of  the  lake. 

The  second  reason  is  the  perfect  freshness  of  the  water. 
This  alone  appears  conclusive  as  to  the  existence  of  an 
outlet,  for  otherwise  the  water  must  gradually  become 
charged  with  salts.  A  similar  case  occurs  in  Berar,  in 
India,  at  the  little  lake  of  Lonar.  This  lake  is  a  little 
smaller,  but  its  drainage  area  is  very  much  less  than  that 
of  Ashangi,  while  the  ground  drained  consists  of  pre- 
cisely similar  rocks.     Lonar  Lake,  however,  instead  of 

^  The  rainfall  in  places  similarly  situated  in  India  varies  from  about  100 
to  200  inches.  Taking  the  drainage  area  of  the  lake  as  a  circle  with  a  radiuB 
of  5  miles,  or  nearly  80  square  miles,  and  estimating  the  area  of  the  lake 
itself  as  12  square  miles,  we  have  an  increase  in  water  for  a  rainfedl  of  80 
inches  of  » <)  ^AQ  =  533  inches,  or  44  feet  6  inches.  Of  this,  if  10  feet, 
an  excessive  estimate  at  this  elevation,  and  in  so  damp  a  dimate,  be  allowed 
for  evaporation^  the  amount  of  increase  would  still  be  more  than  34  feet.  I 
believe  this  estimate  to  be  below  the  truth. 


160  OEOLOOr. 

being  fresh,  is  at  all  times  highly  saline,  and  in  the  hot 
season  it  becomes  a  concentrated  solution,  depositing 
crystals  of  carbonate  of  soda.  It  also  contains  other 
salts. in  large  quantitiea 

The  outlet  from  Ashangi  is  doubtless  to  the  eastward  : 
the  water,  in  all  probability,  percolates  through  the  traps 
and  re-emerges  on  the  flanks  of  the  deep  valley  inha- 
bited by  the  Asubo  Gallas.^ 

It  is  very  difficult  to  account  for  the  formation  of  such 
a  hollow  as  that  of  Ashangi  by  any  known  process  of 
denudation.  It  is  scarcely  credible  that  it  can  be  an 
area  of  depression  ;  the  higher  series  of  traps  capping 
the  lulls  around  are  remarkable  for  the  absence  of  all 
evidence  of  disturbance :  and  it  is  difficult  to  conceive 
the  formation  of  a  small  deep  lake  by  gradual  or  rapid 
sutog  wi4out  ^ppori^g  a.U  to  be  JnTone  in  a  seL 
of  disturbances  which  would  leave  unmistakeable  traces 
on  the  formations  around.  It  is,  however,  certain, 
that  the  lower  group  of  traps  ^  near  Ashangi  is  much 
disturbed,  though  not  more  so  than  it  is  in  many  other 
places,  where  the  upper  group  is  perfectly  horizontal. 

There  is  not  the  smallest  trace  of  glacier  action,  and  it 
may  fairly  be  inferred  that  the  denudation  which  would 
destroy  all  traces  of  glaciers  would  also  fill  up  the  lake 

^  The  valley  immediately  south  of  Ashangi,  in  which  stood  the  camp  of 
Masagita,  appeared  to  me  to  be  equally  without  an  outlet.  It  was  impossible 
to  examine  it,  howeyer,  to  the  eastward,  and  other  observers  thought  differ- 
ently. But  I  cannot  help  thinking  it  possible  that  the  accumulation  of  water 
in  this  hollow  during  the  rainy  season  may  have  occasioned  the  representation 
on  all  the  old  maps  of  a  small  lake  immediately  south  of  AshangL 

'  For  the  distinction  between  these  two,  see  pasted,. 


PHYSICAL  GEOGRAPHY  OF  ABYSSINIA,  161 

with  detritus.  It  is  evident  that  the  lake  must  be  of 
very  small  geological  antiquity,  or  it  would  have  been 
filled  up  long  since.  At  the  same  time  basaltic  rocks 
disintegrate  as  a  rule  much  more  rapidly  on  the  surface 
than  gneiss  and  granite,  and  would  therefore  retain  traces 
of  glacier  action  for  a  less  time.  But  though,  if  the 
whole  globe  was  at  a  much  lower  temperature  during 
the  glacial  epoch,  there  appears  no  improbability  in  the 
existence  of  glaciei-s  even  in  the  tropics  at  a  height  of 
eight  thousand  feet  above  the  sea,^  still  the  ravines 
around  Ashangi  are  so  unlike  glacier  valleys  that  it  is 
scarcely  credible  that  the  lake  can  owe  its  origin  to  the 
action  of  ice. 


^  Tbeir  traces  have  been  found  by  Mr.  H.  B.  Medllcott  in  the  Western 
Uimolayas,  far  below  this,  only  a  few  degrees  further  north. 


M 


162  GEOLOGY. 


SECTION  ir. 

LIST  OF  GEOLOGICAL  FORMATIONS. 

The  various  rocks  observed  in  Abyssinia  and  described 
in  the  subsequent  pages  are  the  following.  They  are 
enumerated  here  in  descending  order,  though,  as  is 
usual  and  most  convenient  in  geological  descriptions, 
they  will  be  treated  in  the  reverse  sequence,  commencing 
with  the  lowest.  Local  names  have  been  given  from  the 
most  important  places  at  which  the  rocks  of  each  group 
{ire  well  developed : — 

6.  Recent  formations — Soils  of  the  highlands,  coral 
islands  of  the  Red  Sea,  alluvial  deposits  near 
the  coast. 

5.  Aden  series  of  volcanic  rocks  bordering  the  Red  Sea. 

4.  Trappean  series  :— 

A.  Ashangi  group. 

B.  Magdala  group. 
3.  Antalo  limestones. 

2.  Adigrat  sandstones. 
1.  Metamorphic  rocks. 

After  passing  the  volcanic  formations  on  the  coast,  the 
route  lay  over  a  simple  ascending  series  of  rocks  as  far  as 


LIST  OF  GEOLOOICAL  FORMATIONS.  163 

Magdala,  the  different  groups  succeeding  each  other  on 
the  road  precisely  as  given  in  the  above  list 

It  should  be  repeated  here  that  no  rock  systems  of 
any  importance  besides  those  above  enumerated  appear 
to  have  been  observed  by  previous  travellers  in  Abyssinia^ 
unless  the  coal-bearing  formations  of  Chelga  belong  to  a 
group  distinct  from  the  Adigrat  sandstone. 


M  2 


1(3 1  GEOLOGY, 


SECTION  III. 


METJ  UORPUICS. 


The  whole  mass  of  the  plateau  of  Northern  Abyssinia, 
as  far  south  as  the  neighbourhood  of  Antalo  at  least, 
consists  of  metamorphic  rocks,  capped  only  by  sedi- 
mentary and  volcanic  formations. 

As  is  usual  in  countries  where  altered  formations  are 
largely  developed,  the  metamorphics  of  Abyssinia  vary 
greatly  in  mineral  character.  The  variation  i^,  indeed, 
greater  than  usual,  every  intermediate  grade  being  found 
between  the  most  coarsely  crystalline  granite  and  a 
slaty  rock  so  little  altered  that  the  lines  of  the  original 
bedding  are  still  apparent. 

Perhaps  the  most  prevalent  form  of  rock  is  a  rather 
finely  crystalline  gneiss.  Hornblende-schist  and  mica- 
schist  are  met  with,  but  neither  of  the  minerals  from 
which  they  are  named  appears  to  be  so  abundant  as  in 
some  metamorphic  tracts.  On  the  other  hand,  a  compact 
felspathic  rock,  approaching  felsite  in  composition,  is 
prevalent  in  places,  as  in  the  Sum  defile,  between 
Komayli  and  Senaf^.  Bands  of  quartzite  were  not  so 
abundant  as  usual  in  the  country  traversed. 


METAMORPHICS.  165 

In  the  region  around  Senafd  and  Adigrat  no  very 
granitoid  tract  was  observed  ;  the  beds  for  the  most  part 
do  not  appear  to  be  highly  crystalline.  Probably  in 
Hamazen,  north  of  Agam^,  the  same  is  the  case,  as  the 
country  is  said  to  be  remarkably  level.  But  at  Keren,  in 
Bogos,  and  again  about  forty  miles  further  to  the  north- 
east, near  Af  Abed  and  Rairo,  in  Habab,  the  rocks  are 
highly  granitoid,  and  even  porphyritic,  containing  large 
crystals  of  felspar.  For  the  most  part  all  traces  of  folia- 
tion have  disappeared,  as  is  usual  in  such  formations,  and 
the  only  structure,  apart  from  the  crystallization,  which 
can  be  detected,  is  the  general  direction  of  the  bands  of 
rock,  often  rising  into  hills.  The  strike  of  these  corre- 
sponds approximately  with  that  of  the  foliation  in  the 
gneiss  of  the  neighbourhood,  and  every  here  and  there, 
throughout  the  area  composed  of  granitoid  rocks,  the 
gneissic  structure  may  be  traced  by  searching  for  it.^ 


*  But  a  few  years  ago  such  granites  as  that  above  described  were  looked 
upon  as  intrusive  rocks,  and  it  was  even  supposed  that  the  alteration  of  the 
metamorphic  fonnations  in  their  neighbourhood  was  due  to  them.  This  view 
is  scarcely  yet  completely  exploded  amongst  geologists,  partly,  perhaps,  be- 
Ciiuse  so  few  have  opportunities  for  seeing  large  areas  of  metamorphic  rocks. 
It  is,  therefore,  well  to  point  out  every  instance  which  cjtn  be  adduced  tending 
to  prove  the  complete  similarity  in  origin  of  the  crystiilline  non-foliated  rocks 
and  of  the  laminated  mica-schist  and  gneiss.  The  appearance  of  the  grani- 
toid tracts  in  Abyssinia  is  precisely  the  same  as  in  India,  where  metiunorphic 
rocks  are  developed  on  the  most  extensive  scale,  and  cover  nearly  half  the 
peninsula  south  of  the  Gangetic  plains.  With  their  great  opportunities  for 
observation,  I  believe  there  is  not  a  single  geologist  on  the  Indian  Survey 
who  questions  the  identity  in  origin  of  the  foliated  and  non-foliated  rocks. 
The  two  pass  into  each  other  everywhere.  The  granitoid  masses  evidently 
owe  their  more  crystalline  condition  either  to  a  diifference  of  chemical  consti- 
tution, original  or  produced  subsequently  to  their  deposition,  which  rendered 
them  better  adapted  to  the  development  of  the  minerals  constituting?  granite 


166  GEOLOGY. 

In  these  granitoid  tracts  the  most  prevalent  rock  is 
composed  of  quartz  and  large  crystals  of  felspar,  mica 
being  present,  but  only  in  very  small  quantities.  The 
hills  have  a  most  peculiar  and  characteristic  appearance. 
They  are  either  huge  crags  with  enormous  cliffs,  the  sur- 
face of  which  is  not  vertical,  but  curved,  or  they  appear 
as  if  composed  of  monstrous  blocks,  more  or  less  rounded, 
and  piled  upon  each  other.  These  peculiar  appearances 
are  due  to  the  mode  of  weathering;  the  rock,  when 
unaltered,  is  so  compact  that  water  can  only  penetrate 
through  joints  and  cracks,  and  it  is  consequently  in  the 
immediate  neighbourhood  of  these  alone  that  disintegra- 
tion takes  place.  As  a  rule,  all  rain-water  runs  off  the 
surface  of  the  rock,  and  consequently  weathering  is  so 
slow  as  to  be  almost  imperceptible  on  tlie  exposed  portion. 
The  accumulation  of  a  little  soil  and  a  few  plants  gene- 
rally produces  a  hollow  on  the  surface,  and  the  rock  in 
contact  with  the  soil  is  usually  seen  to  have  undergone 
change.  The  result  is  that,  while  the  exposed  portions 
of  the  rock  remain  unaltered,  the  surfaces  of  the  joints 
and  cracks  are  gradually  disintegrated  and  washed  away, 
until  the  blocks,  formerly  separated  by  minute  fissures, 
become  entirely  isolated  and  detached.  In  this  posi- 
tion, so  long  as  they  arc  exposed  on  all  sides,  they  are 
scarcely  susceptible  of  change  above,  but  disintegrate 
slowly  beneath,  where  they  are  in  contact  with  soil 
and  vegetation. 

and  allied  rodcs,  or  to  local  conditions  during  the  period  of  metamorphism. 
As  to  gnmite  of  this  kind  ever  having  been  fluid,  certainly  such  evidence  as 
I  have  seen  daring  several  years'  study  of  the  rocks  is  opposed  to  any  such 
supposition. 


METAMORPUICS.  167 

In  ascending  the  mountains  of  Tigr^  from  Komayli  to 
Senafd,  a  gradual  decrease  in  the  amount  of  alteration 
whicli  the  l)eds  had  undergone  is  very  perceptible. 
About  the  lower  portion  of  the  pass,  near  Sum  especially, 
the  rocks  are  much  hardened  and  changed;  but  above 
they  are  slaty,  and  in  places  scarcely  crystalline  at  all. 
Some  beds  in  the  neighbourhood  of  Senafd  are  true  clay 
slates;  the  original  bedding,  in  one  instance  at  least, 
being  still  minutely  visible.  At  first  these  phenomena 
appear  to  indicate  a  gradual  diminution  in  the  amount 
of  alteration  corresponding  with  the  smaller  depth  of  the 
rocks  from  the  surface ;  for  it  is  evident  that  the  deep 
valleys  on  the  flanks  of  the  Abyssinian  mountains  have 
been  cut  out  from  the  plateau  by  running  water,  so  as  to 
expose  rocks  formerly  at  a  depth  at  least  as  great  as  that 
of  the  bottom  of  the  valleys  below  the  high  ground  of 
the  plateau.  But  it  appears  most  probable  that  the  phe- 
nomenon is  merely  local.  The  high  ground  near  Halai 
consists  of  rocks  which  are  more  crystalline  than  near 
Senafd,  and  the  most  crystalline  rocks  seen,  those  above 
referred  to  as  occurring  near  Keren  and  Af  Abed,  are  at 
elevations  of  from  3,000  to  5,000  feet  above  the  sea. 

The  foliation  of  the  metamorphic  rocks  in  Tigr^,  so  far 
as  they  were  observed,  is  remarkably  constant  in  direc- 
tion ;  its  strike  rarely  deviates  more  than  a  few  degrees 
from  north  and  south.  The  dip  is  more  variable,  though 
it  is  generally  nearly  vertical  on  the  highlands ;  near  the 
base  of  the  hills  about  Komayli  and  Hadoda^  and  again 
further  to  the  north,  about  Ain,  it  is  less  than  30^  The 
north  and  south  strike  was  observed  to  prevail  from  the 


168  GEOLOGY, 

neighlx)urhood  of  Ain  and  Af  Abed,  in  Habab  (lat.  15°  55' 
to  16^10'  N.),  as  far  south  as  the  metamorphic  rocks 
were  found  to  predominate  on  the  highlands,  and  that  is 
to  about  twenty-five  niiles  south  of  Adigrat  (or  nearly 
13"*  50'  N.),  or  throughout  a  distance  of  upwards  of  150 
miles  from  north  to  south.  In  the  Anseba  valley,  and 
around  Keren,  in  Bogos,  the  strike  changes  to  north-east 
and  south-west;  and  very  possibly  this  direction  may 
prevail  to  the  northward. 

To  the  constancy  of  the  strike  of  the  metamorphic 
rocks  several  of  the  most  important  features  are  due,  one 
of  the  principal  being  the  direction  of  the  river  valleys. 
No  one  can  have  studied  the  geographical  sketches  made 
during  the  progress  of  the  expedition,  of  the  physical 
features  of  the  country  around  the  passes,  without  ob- 
serving the  remarkable  manner  in  which  the  principal 
valleys,  such  as  those  of  Komayli  and  the  Haddas,  run 
for  many  miles  in  a  north  and  south  direction,  parallel 
to  the  base  of  the  hills,  before  debouching  into  the  plain. 
As  the  facility  with  which  an  excellent  military  road 
was  made  to  the  highlands  was  due  mainly  to  the  length, 
and  consequently  gentle  slope,  of  the  ravines  traversed,  it 
is  very  evident  that  the  accessibility  of  Abyssinia  from 
the  sea  depends,  in  a  great  measure,  upon  the  geological 
structure  of  the  rocks,  and  that  the  foliation  of  the  gneiss 
was  one  of  the  best  friends  of  the  British  army  during  its 
ascent  to  the  highlands. 

Reference  was  made  on  the  previous  page  to  an  in- 
stance being  met  with,  at  the  base  of  one  of  the  hills  near 
Seiuife,  of  the  original  bedding  being  still  perceptible  in 


METAMORPUICS.  169 

a  cleaved  slaty  rock  belonging  to  the  metamorphic  series. 
This  observation  was  of  much  importance ;  for,  as  the 
direction  of  the  planes  of  cleavage  is  the  same  as  that  of 
the  lamination,  viz.  north  and  south  and  vertical,  it  is 
evident  that  the  foliation  in  this  case  corresponds  with 
pre-existing  cleavage.  It  is  also  supported  by  nume- 
rous instances  observable  in  the  slaty  rocks  of  the 
higher  part  of  the  pass  and  elsewhere,  in  which  the 
character  of  the  planes  of  division  is  intermediate 
between  foliation  and  cleavage.^  It  is,  therefore,  highly 
prolmble,  that  the  high  dip,  approaching  verticality,  met 
with  on  the  highlands  is  normal,  and  that  the  low  dips 
at  the  base  of  the  hills  are  due  to  subi^equent  disturbance. 
This  disturbance  may  be  of  late  dat€,  and  synchronous 
with  that  Avhich  has  aftected  the  volcanic  beds  of  the 
coast,  which  \\\\\  be  described  in  a  subsequent  section. 

*  In  a  report  on  the  geology  of  the  lower  Nerbudda  valley  in  the  weBtem 
part  of  the  Indian  ])enin8ula  (Mem.  Geol.  Survey  of  India,  vol.  vi.  p.  193),  I 
have  shown  the  strong  ])robabilities  which  exist  in  favour  of  gneissic  lamina- 
tion corresponding  with  pre-existing  cleavage  whenever  the  dip  is  high  and 
the  strike  constant  over  large  areas.  In  the  Nerbudda  area  the  probability 
of  the  direction  being  due  to  cleavage  was  shown  by  independent  evidence. 
The  metaniorphics  of  Abyssinia  offer  another  instance  to  the  same  effect,  and 
the  rule  may  be  uuiversiil.  It  will  be  of  considerable  geological  importance 
if  this  be  established  by  further  observations  ;  for  if,  as  some  of  our  best 
geologists  believe,  lamination  in  metamorphic  rocks  may  owe  its  direction  to 
either  bedding  or  cleavage,  according  as  the  one  or  the  other  was  the  ruling 
Btnicture  in  the  rock  before  meUimorphism,  the  regularity  and  dip  of  the 
laminatirm  will  aflbrd  a  test  as  to  which  of  the  two  the  present  direction  of  the 
lamination  pluncs  should  be  ascribed. 


1 70  OEOLOOY. 


SECTION  IV. 


ADIQRAT  SANDSTONES, 


In  the  north-eastern  portion  of  Tigrd,  around  Halai, 
Senaf^  and  Adigrat,  the  rock  resting  immediately  upon 
the  metamorphics  is  generally  a  massive  sandstone.  The 
perpendicular  scarps  which  usually  bound  the  areas 
formed  of  this  rock,  and  flank  all  the  valleys  cut  out 
of  it,  are  very  conspicuous,  from  their  height  and  their 
white  colour,  and  they  give  a  very  peculiar  character  to 
the  scenery. 

This  sandstone — which,  on  account  of  its  extensive 
development  around  the  important  town  of  Adigrat,  I 
would  propose  to  call  the  Adigrat  sandstone — is  usually 
white  or  brown  in  colour,  the  former  much  predomi- 
nating. Occasionally  it  is  pale  brown  and  lilac,  or,  as 
about  Takonda,  brick  red ;  these  colours  being  chiefly 
restricted,  however,  to  bands  interstratified  with  the 
mass  of  the  rock.  It  is  usually  very  quaitzose,  fre- 
quently felspathic,  less  commonly  argillaceous.  Shales 
of  a  blue  or  lilac  colour  are  frequently  met  with  towards 
the  base  of  the  group,  but  the  principal  characteristic  of 
the  great  bulk  of  the  sandstone  is  its  massive  character, 
and  the  absence  of  marked  bedding,  so  that  the  high 


JDIQRAT  SANDSTONES.  171 

cliffs  of  it  which  form  the  head  of  the  Haddas  and 
Komayli  ravines,  and  wliich  surround  valleys  such  as 
Baragit  and  Guna  Guna,  south  of  Senafd,  appear  as  if 
cut  out  of  a  huge  unstratified  block. 

In  thickness  the  sandstone  varies ;  and  as  the  original 
surface  is  rarely,  if  ever,  seen,  only  an  approximate  idea 
can  be  formed.  On  Sowera  it  is,  by  aneroid  measure- 
ment, about  1,000  feet,  or  perhaps  rather  more,  as  the 
observation  was  not  taken  on  the  highest  peak.  About 
Guna  Guna  it  does  not  much  exceed  half  that  thickness. 
At  Adigrat  it  appears  to  be  intermediate  in  amount. 

No  fossils  were  met  with  in  this  group,  although 
portions  of  it,  the  shales  especially,  were  much  searched 
for  organic  remains.  This  is  unfortunate,  for  its  position 
in  the  geological  series  can  only  be  approximately  deter- 
mined. By  Messrs.  Ferret  and  Galinier  it  was  classed, 
with  doubt,  as  tertiary.  There  can  be  but  little  doubt 
that  it  is  much  older  than  this,  from  its  relations  to  the 
Antalo  limestone ;  but  its  true  age  is  still  uncertain. 

As  above  remarked,  it  invariably  rests  upon  the  meta- 
morpliics ;  no  other  rocks  have  been  observed  to  inter- 
vene. In  the  neighbourhood  of  Senaf^  and  Adigrat,  the 
only  rocks  seen  resting  upon  it  are  traps,  which  are 
apparently  conformable  at  the  latter  place,  but  quite 
unconformable  at  the  former.  Their  relations  to  the 
sandstone  will  be  described  further  on. 

About  twenty-five  miles  south  of  Adigrat,  at  Mai 
Dongolo  or  Wugula,^  sandstones  are  seen  to  dip  below 

J  Not  Agiik,  the  name  given  to  another  camp,  which  is  ten  miles  further 
south. 


172  GEOLOGY. 

the  Antalo  limestone.  The  Aclignit  sandstone  is  traced, 
almost  without  a  break,  from  the  neighbourhood  of 
Halai,  Takonda,  and  Senaf(5,  to  within  about  six  miles 
east  of  Dongolo ;  but  then  there  is  a  break,  and  meta- 
morphics  intervene,  so  that  the  beds  at  Dongolo  are  not 
connected  with  the  mass  of  the  formation.  They  diflfer 
also  in  colour,  being  red ;  and  instead  of  being  horizontal, 
as  at  Adigrat  and  generally  throughout  Northern  Tigre, 
they  dip  at  a  considerable  angle  to  the  southward. 

These  differences  do  not  appear  to  be  of  much  im- 
portance. The  colour  of  the  sandstone  is  occasionally 
red  everywhere,  and  is  very  liable  to  vary ;  and  although 
the  greater  portion  of  the  formation  is  perfectly  hori- 
zontal, it  does  dip  in  places.  Besides,  the  Antalo  lime- 
stones are  also  horizontal  in  general,  although  they  too 
are  considerably  inclined  near  Dongolo.  Moreover,  the 
sandstone,  a  little  farther  west  than  Dongolo,  forms 
massive  craggy  hills,  standing  out  above  the  plain  of 
the  metamorphics,  which  is  higher  than  the  valley  in 
which  Dongolo  lies ;  and  these  crags,  although  formed 
of  inclined  beds,  resemble  closely  those  of  the  Adigrat 
sandstone  further  to  the  eastward  and  northward ;  in- 
deed, they   have   all  the  appearance  of  being  outliers 


Sketch  section  of  relations  between  trap,  limestone,  sandstono,  and  metamorphics 
near  Dongolo  :  a,  trap  ;  ft,  limestone  ;  r,  sandstone  ;  d,  metamorj^hios. 

from  the  Adigrat  hills.     The  accompanying  sketch  sec- 
tion will  give  some  idea  of  the  apparent  relations  of  the 


JDIGRA  T  SANDSTONES.  173 

beds  near  Dongolo,  the  vertical  dimensions  being,  of 
course,  greatly  exaggerated.  The  mineral  character  of 
the  sandstone  at  Dongolo  is  precisely  similar  to  that  of 
the  Adigrat  beds  in  general. 

No  trace  of  the  sandstones  was  o1)scrved  further  south 
on  the  road  to  Magdala ;  all  the  beds  seen  were  higher  in 
the  series,  except  at  one  spot,  near  the  camp  at  Dolo, 
where  a  small  boss  of  granite  emerges  from  beneath  the 
limestone.  This  small  patch  is  so  minute  that  it  aflFords 
no  means  of  judging  whether  the  sandstones  to  the 
southward  underlie  the  limestone  or  not.  Both  forma- 
tions are,  of  course,  quite  unconformable  to  the  meta- 
morphics,  and  rest  upon  an  uneven  surface  of  them,  so 
that  peaks  of  the  latter  may  rise  through  all  the  former 
sedimentary  formations,  and  only  be  capped  by  the 
highest. 

Piissing  northward  from  their  southern  boundaiy  at 
Dongolo,  the  sandstones,  as  already  intimated,  cover  a 
considerable  tract  around  Adigrat,  and  form  the  lower 
portion  of  the  great  hill-range  west  of  that  town  ;  thence, 
to  the  north,  they  cross  the  highest  portion  of  the 
dividing  range  between  the  streams  running  down  to 
the  Nile  basin  on  the  west,  and  those  flowing  to  the  salt 
plain  west  of  Amphila  to  the  east.  Their  area  is  very 
narrow  as  far  as  Senafe,  but  here  they  expand  into  two 
or  three  broader  plateaux,  which  encircle  the  upper  por- 
tions of  the  Komayli  and  Haddas  valleys.  The  most 
eastern  of  these  plateaux  is  Sowera,  north-east  of  Senafd. 
Further  still  to  the  north-east  and  east,  sandstone  caps 
some  of   tlie  isolated  rid<]:es  ninninij  north  and  south 


1 74  QSOLOQr. 

l)etween  the  highlands  and  the  plains  near  Annesley  Bay. 
One  such  cap  is  seen  north-east  of  Sowera  on  the  oppo- 
site side  of  a  deep  ravine.  Another  lies  east-south-east 
of  Sowera.  A  third  is  not  seen  from  the  highlands,  but 
was  found  at  a  much  lower  level  in  the  hills  further  east 
towards  Arafild  by  Lieutenant  Holditch,  R.E.,  of  the 
Trigonometrical  Survey. 

The  whole  of  the  Sowera  plateau  consists  of  sand- 
stone, but  the  rock  ends  somewhat  abruptly  on  the 
Asawat  plateau  between  the  Haddas  and  Komayli  ravines, 
and  also  west  of  the  former,  near  Halai,  which  village 
stands  on  metamorphic  rocks.  It  is  curious  that  the 
northern  part  of  each  plateau,  which  consists  of  meta- 
morphics,  is  higher  than  the  southern  part.  This  shows 
on  how  uneven  a  surface  the  sandstone  was  deposited. 
The  limit  of  that  formation  nearly  corresponds  with  an 
east  and  west  line  on  both  sides  of  the  Haddas  valley, 
but  there  does  not  appear  good  evidence  of  faulting. 
On  the  plateau  of  Hamazen,  north  of  Halai,  there  is  said 
to  be  no  sandstone. 

The  sandstone  plateaux  are  covered  with  thin  jungle, 
and  do  not  appear  to  be  suflSciently  fertile  for  cultivation 
in  general,  except  where  a  richer  soil  is  derived  from 
trap  rocks  in  the  neighbourhood,  as  is  the  case  near 
Adigrat. 

The  extent  of  the  sandstone  to  the  westward  was  of 
course  not  determined.  Looking  from  the  verge  of  the 
scarps  near  Takonda^  it  appears  that  for  some  distance 
metamorphics  prevail,  but  that  the  sedimentary  rocks 
recur  on  some  ranges  beyond  Dixa. 


ADIGRAT  SANDSTONES,  1 75 

It  is  possible  that  the  beds  associated  with  coal  in  the 
Chelga  district  may  be  of  the  same  age  as  the  Adigrat 
sandstone.  It  is  very  much  to  be  regretted  that  the 
problem  of  their  age,  one  of  the  most  interesting  geo- 
logical questions  in  Abyssinia,  could  not  be  solved.^ 

1  I  would  venture  to  suggest  the  possibility  that  both  the  coal-beariog  beds 
of  Chelga  and  the  Adigrat  sandstones  may  belong  to  a  portion  of  the  great 
series  associated  with  coal  in  India — the  Damudas  and  their  associates  of  the 
Indian  Survey  classification.  This,  of  course,  is  mere  surmise,  and  my  only 
reason  for  thinking  it  probable  is  the  resemblance  between  the  fossils  of  the 
dicynodont  beds  of  South  Africa  and  the  Panchets  of  India,  and  the  want^  so 
far  as  is  known,  in  Equatorial  and  Southern  Africa,  as  in  great  part  of 
Southern  Asia,  of  marine  representatives  of  the  upper  palaeozoics  and  lower 
mesozoics  of  Europe. 


1 76  OEOLOOr, 


SECTION  V. 


ANTALO  LIMESTONES, 


Next  above  the  sandstone  in  ascending  order  is  a  thick 
group  of  limestones,  which  rest  upon  the  Adigrat  beds 
at  Mai  Dongolo  in  the  manner  described  in  the  last 
section,  and  thence  cover  the  country  for  a  distance  of 
seventy  miles  along  the  route,  as  far  as  about  twenty 
miles  south  of  Antalo.  They  extend  also  for  a  very 
considerable  distance  east  and  west. 

For  twenty  miles  south  of  Dongolo  the  limestone  is 
not  accompanied  by  any  other  rock.  It  is  usually  in 
thin  beds  of  a  grey  colour,  less  commonly  ochreous, 
and  it  much  resembles  some  of  the  beds  of  lias  lime- 
stone in  the  south-west  of  England.  The  rock,  when 
broken,  is  compact  and  earthy,  or  but  slightly  crystalline 
in  general. 

Further  south,  towards  Antalo,  although  limestone  is 
still  the  prevailing  rock,  some  beds  of  sandstone  and  of 
trap  are  interstratified  ;  the  former  especially  prevailing 
towards  the  upper  part  of  the  group.  The  hill  of  Gutba 
Hariat,  behind  Antalo  town,  appears  to  be  in  great 
measure  composed  of  sandstone ;  and  so  are  the  beds 
immediately  beneath  the  trap  in  the  Meshek  valley,  south 


ANTALO  LIMESTONES.  1  7  7 

of  Antalo.  These  sandstones  differ  in  character  from 
the  Adigrat  beds.  They  are  less  quartzose,  being  both 
more  earthy  in  composition  and  darker  in  colour,  and 
they  contain  large  quantities  of  conglomerate.^ 

The  trap  is  always  doleritic,  and  usually  compact 
basalt.  Beds  of  it  are  seen  at  Mai  Makdam,  between 
Agula  and  Dolo,  but  they  are  irregular,  and  have  rather 
the  appearance  of  outliers  of  a  higher  formation,  depo- 
sited on  an  irregular  surface  of  limestone,  than  of  inter- 
stratified  lava  flows.  Farther  south,  however,  and 
especially  between  the  camp  at  Buya,  near  Antalo,  and 
Meshek,  there  can  be  no  question  of  the  basalts  being 
contemporaneous.  One  bed  at  Masgi,  the  first  camping- 
ground  south  of  Antalo,  is  about  one  hundred  feet  thick, 
as  nearly  as  could  be  measured  with  an  aneroid  on  the 
hill-side. 

As  a  rule,  the  Antalo  limestone  is  horizontal.  It  does 
not,  however,  preserve  the  same  unbroken  lines  as  the 
sandstone  near  Adigrat,  but  rolls  about  slightly  every- 
where.    In  places  it  is  much  disturbed,  as  at  Meshek. 

The  scener}^  is  peculiar.     The  limestone  forms  cliffs,  of 

1  There  is,  of  course,  the  bare  possibility  that  these  sandstones  at  the  top 
of  the  Antalo  group,  and  not  those  of  Dongolo,  represent  the  Adigrat  beds. 
For  the  reasons  given  in  the  text,  I  think  this  highly  improbable.  The  sand- 
stones of  Antalo  and  Meshek  have  every  appearance  of  being  an  integral 
portion  of  the  Antalo  beds,  and  limestone  more  or  less  impure  is  mixed  with 
them.  No  limestone  whatever  occurs  either  at  Dongolo  or  in  the  Adigrat 
beds.  The  only  reason  I  am  acquainted  with  for  supposing  the  limestone  to 
be  of  higher  antiquity  than  the  sandstone  of  Adigrat  is,  that  the  former  is  less 
persistently  horizontal  than  the  latter,  the  beds  usually  rolling  about  slightly. 
This  may  be  due  to  local  causes,  but  it  is  almost  universally  the  case  in  thick 
groups  of  beds,  when  the  strata  themselves  are  not  massive,  and  especially  so 
amongst  limestones. 


1 78  GEOLOGY, 

no  great  height  in  general,  along  the  sides  of  the  valleys, 
all  of  which  contain  streams,  and  have  a  striking  cha- 
racter, due  to  the  nearly  perpendicular  sides  and  the 
narrow  flats  covered  with  grass  and  bushes  at  the 
bottom.  Except  in  the  valleys,  the  soil  is  poor,  and 
only  low  scattered  scrub  occurs  on  the  hills.  The  latter 
^ire  a  rolling  upland,  very  distinct  from  the  plateaux  of 
sandstone  or  the  terraced  ridges  and  rounded  hummocks 
of  the  trap. 

Where  the  Antalo  group  rests  upon  the  Adigrat 
sandstones  at  Mai  Dongolo,  the  two  formations  are 
apparently  conformable.  Whether  they  are  so  through- 
out it  is  impossible  to  say;  there  was  not  time  for 
sufficient  exploration  of  the  country  to  determine  the 
question.  The  non-occurrence  of  any  limestone  upon 
the  sandstones  of  Adigrat  and  Senafd  is  opposed  to  the 
idea  of  the  two  being  conformable.  A  similar  doubt 
rests  upon  the  relations  of  the  lower  series  of  traps  with 
the  limestone,  as  will  be  seen  in  treating  of  the  volcanic 
rocks  in  the  next  section. 

This  want  of  opportunity  for  determining  precisely 
the  relations  of  the  limestone  to  the  rocks  above  and 
below  is  the  more  to  be  regretted,  as  the  present  is  the 
only  group  which  has  yielded  fossils.  Organic  remains 
abound  in  it,  as  so  frequently  occurs  amongst  limestone ; 
but  in  general  they  consist  only  of  oysters,  half-defaced 
casts  of  bivalves,  and  spines  of  Echinodermatay  none 
of  which  are  of  much  value  for  affording  determinations 
of  age.  For  a  long  time  it  seemed  to  me  as  if  similar 
ill-preserved  specimens  were  the  only  fossils  to  be  found, 


ANTALO  LIMESTONES,  1 79 

and  during  the  huiTied  march  to  Magdala  nothing  of 
any  value  was  obtained  ;  but  when  returning  it  was 
possible  to  devote  a  couple  of  days  to  examining  the 
neighbourhood  of  Agula  and  Dongolo,  and  the  search 
was  rewarded  by  the  discovery  of  a  few  species  in  better 
condition.     Of  these  the  following  is  a  list '} — 

ECHINODERMATA — 

f  Hemicidaris  Ahyssinica,  sp.  nov. 
Cidari^,  ?  sp.  a  fragment. 

MoLLUsoA  Lahellibrakchiata — 

Ostrea,  sp.  near  0.  Jonesiana,  Tate. 

GryphcEay  sp. 

Pfjcten,  sp. 

Trigonia  costata,  Sow.  yar.  pullus, 

Modiola  Baiiii,  ?  Sharpe. 

Modiola  irnbricarxa,  Sow.  var. 
f  MyiU'os  Tigrensist  sp.  nor. 

Cyprina,  sp.  cast. 

Tancredia,  sp. 

Isocardia,  ?sp.  cast. 

TeUina,  perhaps  two  species,  casts. 

Ceromya  concentrica,  Sow. 

C.  similiSf  Sow. 
f  C.  paucilirata,  sp.  nov. 

f  Pholadomya  sublirata,  sp.  nov.   near  P.  recurvti,  Ag.,  and 
P.  concatenata,  Ag. 

P.  sp.  near  P.  antica,  Ag. 
f  P.  fframdifera,  sp.  nov. 

Gasteropoda — 
Natica,  pp.  casts. 
Alaria,  sp.  do. 
Cerithium,  sp.  fragmentary. 

(f  For  the  descriptions  of  the  new  species,  see  Appendix  to  this  part.) 

The  species  of  Ilemicidarisy  Plwladomya,  Ceromya, 
Trigonia,  and  Alaria  are  all  of  characteristically  oolitic 

1  I  am  greatly  indebted  to  my  friends  Dr.  Stoliczka  and  Mr.  Etheredge  for 
aid  in  determining  these  fossils. 

N2 


180  OEOLOOY. 

forms,  and  amply  suflSce  to  determine  the  series  to  which 
these  limestones  belong.  The  fauna  has  a  general  resem- 
blance to  that  of  the  middle  Jurassics,  but  the  evidence 
is  scarcely  suflScient  to  justify  any  decided  conclusions  in 
matters  of  detail 

The  occurrence  of  trap  beds  interstratified  with  the 
limestones  is  very  interesting,  because  it  indicates  a  com- 
mencement, in  the  oolitic  period,  of  the  great  series  of 
volcanic  eflFusions,  the  remains  of  which  still  cap  so  many 
of  the  Abyssinian  mountains.  It  also  shows  that  the 
earliest  lava-flows  must  in  all  probability  have  been 
submarine,  like  the  limestones  with  which  they  are 
intercalated.  It  is  worthy  of  note  that  the  earliest  great 
outburst  of  traps  of  known  age  in  India  is  also  Jurassic. 


SECTION  VI. 


TRAPPEAN  SERIES. 


The  greater  portion  of  the  Abyssinian  plateau,  with 
the  exception  of  its  north-eastern  comer,  is  known  \o  be 
covered  by  volcanic  rocks.  Much  has  yet  to  be  done 
before  their  geology  can  be  properly  ascertained.  No 
opportunity  was  aflforded  during  the  expedition  of  visit- 
ing the  remarkable  trachytic  ranges  around  Adowa,  and 
it  can  only  be  inferred  from  the  form  of  the  hills  that  the 
rocks  are  similar  to  those  of  Senaf^.  Far  to  the  south  in 
Shoa  the  trap  rocks,  as  described  by  M.  Rocher  d'Hdri- 
court,^  appear  to  correspond  with  those  of  the  neigh- 
bourhood of  Magdala.  From  the  accounts  given  by 
Messrs.  Ferret  and  Galinier,  Dr.  Ruppell,  and  others,  it  is 
evident  that  the  high  mountains  of  Simcn  and  the  whole 
country  around  Lake  Dembea  consist  of  the  same  bedded 
traps.  By  the  kindness  of  Dr.  Beke  I  have  been  allowed 
to  examine  a  series  of  specimens  collected  during  his 
journeys  in  Central  Abyssinia,  which  entirely  confirm 
these  accounts,  and  prove  that  the  rocks  of  the  hills  on 
the  banks  of  the  Abai,  and  those  both  north  and  south  of 
Lake  Dembea,  are  precisely  similar  to  the  traps  of  Lasta. 

^  Bull,  de  la  Soc.  G4ol.  de  France,  1846,  p.  543. 


182  GEOLOGY, 

On  the  route  to  Magdala  volcanic  rocks  were  first  met 
with  at  Senafd,  where  several  hills  consist  of  trachyte 
passing  into  claystone  and  basalt.  Trap  hills,  chiefly  of 
trachyte,  are  dotted  over  the  country  to  the  southward  as 
far  as  Fokada,  a  distance  of  nearly  thirty  miles.  Here  a 
great  range  of  bedded  traps  commences,  and  extends  for 
about  twenty-five  miles  to  the  south,  passing  to  the  west 
of  Adigrat.  From  a  little  south  of  that  town  no  volcanic 
rocks  were  met  with,  except  the  beds  mentioned  in  the 
last  section  as  occurring  interstratified  with  the  Antalo 
limestones,  imtil  some  distance  south  of  Antalo  ;  but  at 
Meshek,  two  marches  beyond  that  town,  the  route  entered 
high  ranges  entirely  composed  of  trap,  and  thence  no 
other  rocks  were  seen  as  far  as  Magdala. 

In  the  Meshek  valley  and  in  other  valleys  to  tlie 
south,  especially  at  Atala,  Aiba,  and  Belago,  it  is  very 
plainly  seen  that  the  trappean  rocks  belong  to  two  dis- 
tinct and  unconformable  groups.  The  lower  of  these  is 
much  inclined,  while  the  higher  rests  on  its  upturned  and 
denuded  edges.  In  this  part  of  the  country  the  two 
groups  are  easily  distinguished  by  the  absence  of  dis- 
turbance in  the  higher  beds.  Nothing  can  be  more 
complete  than  the  evidence  of  unconformity.  Of  these 
two  groups  the  lower  or  inclined  beds  may,  for  the  sake 
of  distinction,  be  called  the  Ashangi  group,  from  their 
development  aroimd  the  lake  of  that  name,  and  the 
higher  or  horizontal  beds  the  Magdala  group.  They  are 
evidently  of  diflferent  geological  age,  and  require  separate 
description. 


TRAP FE AN  SERIES.  183 

A.   ASHANGI  GROUP   OF   TRAPS. 

The  lower  group  consists  entirely  of  doleritic  rocks. 
The  basalts,  of  which  it  is  principally  composed,  are 
usually  amygdaloidal,  containing  nodules  of  agate  and 
zeolite,  frequently  coated  with  green  earth.  The  zeolite 
is  usually  white  or  orange-coloured  stilbite.  Beds  of 
volcanic  ash  or  breccia  frequently  occur,  and  often  contain 
augite  crystals.  Not  unfrequently  the  rocks  composing 
this  group  are  much  decomposed,  and  have  a  red  colour, 
a  sure  mark  in  general  amongst  basaltic  rocks  of  alteration 
by  atmospheric  action.     Some  beds  are  scoriaceous. 

At  Atala  and  Aiba  these  beds  are  only  seen  in  the 
bottom  of  the  valleys,  but  from  the  Ferra  pass  near 
Belago  they  extend  far  to  the  south,  the  greater  part  of 
the  hills  around  Lake  Ashangi  being  composed  of  them, 
and  probably  all  the  country  thence  to  the  southward  as 
far  as  the  upper  Tellari  valley.  The  distinction  between 
the  two  groups  was  not  made  out  in  time  to  allow  of 
their  being  separated  further  to  the  south,  and  it  is  im- 
possible to  say  now  whether  the  beds  exposed  in  the 
Jitta  and  Bashilo  valleys  belong  to  the  present  division  or 
not.  There  can  be  but  little  doubt  that  the  rocks  seen  in 
the  bottom  of  the  upper  Takkazzy6  valley  are  a  part  of 
the  Ashangi  group,  and  probably  the  same  are  exposed  in 
the  much  deeper  gorges  near  Magdala.  They  have  not, 
however,  been  represented  as  occurring  on  the  map  south 
of  the  Tellari,  and  their  boundaries  throughout  are  merely 
approximate. 

The  only  place  where  the  Ashangi  traps  were  seen  to 


184  GEOLOGY. 

rest  on  the  lower  rocks  was  in  the  Meshek  valley.  Here 
the  geology  is  greatly  confused,  the  Antalo  limestones  and 
their  associates  being  much  disturbed  and  contorted. 
Above  them  are  seen  inclined  basaltic  beds  belonging  to 
the  Ashangi  group,  capped  by  the  horizontal  Magdala 
traps.  The  Ashangi  beds  appeared  to  be  imconformable 
upon  the  sandstones  and  limestones  beneath  them,  but 
the  intricacy  of  the  geology  was  too  great,  in  a  country 
much  covered  by  bush  and  tree  jungle,  to  enable  this  to 
be  determined  with  certainty  in  a  single  morning,  all  the 
time  that  could  be  devoted  to  its  examination.  A 
priori^  there  is  of  course  a  probability  that  these  inclined 
beds  might  belong  to  the  same  series  as  the  limestones,  in 
which,  as  has  been  shown,  beds  of  basalt  are  interstra- 
tified,  especially  towards  the  top ;  but  the  sections  seen 
at  Meshek  were  opposed  to  this  view.  Still,  it  cannot 
be  considered  as  absolutely  disproved. 

Nothing,  however,  can  be  considered  as  ascertained 
concerning  the  age  of  these  traps,  except  that  they  are 
not  older  than  oolitic.  In  mineral  character,  they 
strikingly  resemble  the  great  trappean  series  of  Western 
India ;  the  resemblance  even  extending  to  such  minute 
peculiarities  as  the  frequent  occurrence  of  agate  or  zeolite 
covered  with  green  earth,  and  the  existence  of  a  peculiar 
porphyritic  basalt  with  tabular  crystals  of  felspar.  The 
age  of  the  Indian  traps  has  of  late  been  shown  to  be 
almost  certainly,  in  great  part  at  least,  upper  cretaceous ; 
but  resemblance  of  mineral  character  is  hardly  sufficient 
alone  to  connect  rocks  in  countries  so  widely  distant 
from  each  other  as  Abyssinia  and   Bombay.     Still,  as 


TRAP P BAN  SERIES.  185 

somewhat  similar  beds  are  known  to  occur  in  places 
along  the  coast  of  Arabia,  underlying  the  nummulitic 
series,  just  as  the  Bombay  traps  do  in  Guzerat  and 
Cutch,  it  is  a  very  interesting  question  for  future  in- 
vestigation, what  connexion  exists  between  the  various 
upper  secondary  or  lower  tertiary  bedded  traps  of 
South-western  Asia  and  Eastern  Africa?  Should  they 
be  proved  to  have  been  formerly  connected,  and  to  be 
portions  of  the  same  great  ancient  volcanic  region,  an 
idea  which  seems  by  no  means  improbable,  their  study 
will  become  one  of  very  great  interest  as  connected  with 
the  geological  history  of  the  earth's  surface.  In  every 
respect  the  bedded  traps  are  very  remarkable,  and  deserv- 
ing of  far  more  attention  from  geologists  than  they  have 
hitherto  received.  Nothing  resembling  them  has  yet 
been  shown  to  be  in  process  of  formation  upon  the 
earth's  present  surface,  and  their  mode  of  origin  is  still 
extremely  obscure.^ 

B.    MAGDALA  GROUP   OF  TRAPS. 

The  higher  volcanic  beds  are  distinguished  petrolo- 
gically  from  the  lower  mainly  by  the  frequent  occurrence 
of  thick  beds  of  trachyte,  usually  more  or  less  crystal- 
line, and  not  unfrequently  containing  small  beautifully- 
formed  glassy  crystals  of  felspar  with  rounded  angles. 

^  I  have  endeavoured,  in  a  paper  on  the  "  Traps  of  Western  India,"  pub- 
lished in  the  "  Memoirs  of  the  Geological  Survey  of  India,"  vol.  vi.,  to  bring 
together  all  that  is  known  as  yet  concerning  the  bedded  traps  of  that 
country.  A  great  part  of  the  description  is  equally  applicable  to  the  lockB 
of  Abyssinia. 


186  GEOLOGY. 

These  beds  usually  weather  with  a  remarkably  pre- 
cipitous scarp,  and  being  of  considerable  thickness  and 
perfectly  horizontal,  they  are  unusually  conspicuous. 
To  them  is  due  in  a  great  measure  the  great  difficulty 
of  access  of  most  of  the  high  ranges,  and  especially  of 
the  ambas  or  hill-forts,  which  are  so  abundant  in  Abys- 
sinia. The  higher  ranges  are  perhaps  as  fine  an  example 
as  could  be  found  of  the  terraced  ridges  which  are  so 
characteristic  of  bedded  traps.  There  is  a  remarkable 
resemblance  to  the  scenery  of  the  Western  Deccan  and  the 
higher  valleys  of  the  western  ghats  in  India,  but  the  pecu- 
liarities of  the  landscape  are  exaggerated  in  Abyssinia. 

Many  of  the  trachjrtic  beds  are  brecciated,  and  several 
are  highly  columnar.  The  dolerites  associated  with  them 
are  usually  compact,  amygdaloid  being  less  abundant 
than  in  the  Ashangi  beds.  Still  it  is  far  from  uncommon, 
and  the  nodules  are  occasionally  enveloped  in  green 
earth.  An  instance  of  this  was  observed  at  the  crest  of 
the  Alaji  pass  between  Meshek  and  Atala. 

Sedimentary  rocks  are  interstratified  with  this  series 
in  places,  as  in  the  gorge  of  the  Jitta  river,  and  again 
near  Magdala.  The  beds  met  with  were  white  argil- 
laceous rock  at  both  places,  and,  at  the  Jitta  alone, 
coarse  sandstone  and  black  shale.  The  sections  have 
been  described  in  the  first  part  of  this  work.  The  beds 
do  not  appear  to  extend  for  any  distance. 

No  fossils  could  be   found  -}   there   is   consequently 

1  I  was  informed  by  Dr.  Schimper  that  when  King  Theodore  made  his 
military  road  for  the  transport  of  his  guns  across  the  Jitta  gorge,  some 
petrified  trunks  of  trees  were  met  with  in  the  sandstone. 


TRAPPEAN  SERIES.  187 

nothing  to  indicate  the  geological  age  of  this  very  in- 
teresting group  of  traps.  No  other  formation  was  seen 
resting  upon  it ;  it  caps  all  the  highest  hills  observed. 

It  is  highly  probable  that  the  trap  rocks  of  Shoa 
belong,  in  part  at  least,  to  this  group.  No  idea  can  be 
formed  of  their  extension  further  south,  nor  is  it  possible 
to  say  whether  the  trappean  formations  which  cover  so 
large  an  area  in  Central  Abyssinia  should  be  referred  to 
the  Magdala  or  the  Ashangi  group.  In  all  probability 
both  are  represented.  There  can  be  but  little  doubt, 
however,  that  the  traps  on  the  hills  west  of  Adigrat: 
belong  to  the  higher  group.  Although  the  lower  beds 
consist  of  basalt  alone — and  from  want  of  time  it  was 
impossible  to  examine  the  higher  hills — the  cliflf-lik(» 
scarps  are  just  as  conspicuous  amongst  them  as  farther 
south ;  and  these  appeared,  wherever  they  could  bo 
examined,  to  be  characteristic  of  the  trachytes.  The 
conformity  to  the  Adigrat  sandstone  is  in  all  probability 
only  apparent. 

The  volcanic  rocks  of  the  neighbourhood  of  Senafe 
diflfer  from  those  of  Adigrat  and  the  south  in  the  absence 
of  any  marked  bedding,  and  to  some  extent  also  in 
mineral  charactc^r.  They  consist  chiefly  of  trachyte, 
but  this  is  frequently  so  fine-grained  and  compact,  that 
even  with  the  aid  of  a  lens  it  is  very  difficult  to  dis- 
tinguish it  from  a  sedimentary  rock,  being  in  fact  what 
is  generally  termed  claystone.  The  crags  just  west  of 
the  camp  at  Senafe  were  most  naturally  mistaken  by 
everybody  in  the  army  for  sandstone.^     These  hills  are 

'  As  already  mentioned,  it  was  some  time  before  I  was  myself  undeceiyed 


188  GEOLOGY, 

rounded  hummocks  of  a  very  peculiar  shape.  Many- 
more  of  similar  form  occur  to  the  westward,  and  the 
range  near  Adowa  presents  a  very  similar  appearance, 
and  consists,  in  all  probability,  of  the  same  rock. 

These  traps  rest  in  places  upon  the  metamorphics, 
but  in  general  upon  the  Adigrat  sandstones,  to  which 
they  are  quite  unconformable.  Although  they  are  not 
generally  bedded,  there  is  in  places,  as  on  Akub  Teriki 
hill,  an  appearance  of  distinct  strata ;  and  the  various 
trachyte  hills  to  the  south,  towards  Guna  Guna  and 
Fokada,  have  much  the  appearance  of  being  the  remnants 
of  one  thick  lava  flow. 

Taking  into  consideration  their  different  mineral 
character  aud  mode  of  accumulation,  the  Adowa  and 
Senafd  traps  have  much  the  appearance  of  belonging  to 
a  distinct  group  from  the  beds  of  Ashangi,  Magdala,  and 
Adigrat,  and  their  much  greater  apparent  unconformity 
to  the  sandstones  beneath  would  favour  the  idea  of  their 
being  more  recent  than  the  rocks  of  the  Adigrat  hills. 
If  they  be  so,  then  their  age  must  be  very  much  less, 
for  there  can  be  but  little  doubt  that  the  traps  of  the 
Adigrat  hills  had  formerly  a  much  greater  horizontal 
extension :  they  are  doubtless  merely  the  remains  oi 
beds  which  once  covered  the  whole  country  to  a  depth 
probably  of  2,000  or  3,000  feet,  and  all  of  these  must 
have  been  swept  away  again  at  Senaf^  before  the  volcanic 
formations  of  Akub  Teriki  and  the  neighbouring  hills 

in  this  matter.  Where  the  rock  is  slightly  decomposed,  as  it  generally  is,  it 
is  almost  impossible  to  distinguish  it  from  an  argillaceous  sandstone.  Dr. 
Riippell  took  these  rocks  for  hardened  marl. 


TRAP PE AN  SERIES,  189 

were  poured  out.  It  appears  more  probable  that  the 
latter  rocks  are  really  contemporaneous,  and  that  their 
irregularity  is  due  to  their  accumulation  in  the  imme- 
diate vicinity  of  active  vents, — to  their  having  been,  in 
short,  part  of  a  volcanic  focus.  The  deep  beds  of  Akub 
Teriki  may  be  the  intrusive  mass  of  the  vent  itself. 
The  remaining  hills,  where  thick  trachyte  rests  on  sand- 
stone, may  be  parts  of  an  old  lava  flow  of  great  deptL 
Near  Fokada  there  is  evidence  in  favour  of  a  passage 
from  the  Scnaf^  traps  into  those  of  the  neighbourhood 
of  Adigrat.  It  therefore  appears  most  probable  that  aU 
belong  to  one  group. 


180  OEOLOOY. 

forms,  and  amply  suflSce  to  determine  the  series  to  which 
these  limestones  belong.  The  fauna  has  a  general  resem- 
blance to  that  of  the  middle  Jurassics,  but  the  evidence 
is  scarcely  suflScient  to  justify  any  decided  conclusions  in 
matters  of  detail. 

The  occurrence  of  trap  beds  interstratified  with  the 
limestones  is  very  interesting,  because  it  indicates  a  com- 
mencement, in  the  oolitic  period,  of  the  great  series  of 
volcanic  eflFusions,  the  remains  of  which  still  cap  so  many 
of  the  Abyssinian  mountains.  It  also  shows  that  the 
earliest  lava-flows  must  in  all  probability  have  been 
submarine,  like  the  limestones  with  which  they  are 
intercalated.  It  is  worthy  of  note  that  the  earliest  great 
outburst  of  traps  of  known  age  in  India  is  also  Jurassic. 


ALES  SERIES  OF  VOLCANIC  ROCKS.  191 

iuterstratified  with  which  are  sedimentary  rocks,  sand- 
stones, and  conglomerates. 

As  a  rule  these  beds  are  inclined,  and  present  the 
appearance  of  having  been  much  disturbed.  The  beds 
of  the  Buri  peninsula  were  only  examined  in  one  spot. 
Those  of  the  neighbourhood  of  Zulla  are  most  irregular 
in  dip,  whilst  in  the  tract  west  and  north-west  of  Mas- 
sowa,  with  some  irregularity,  there  is  a  very  constant 
general  dip  towards  the  sea  to  the  eastward. 

The  sedimentary  beds  contain  a  mixture  of  sand,  ashes, 
and  rolled  fragments  of  both  volcanic  and  metamorphic 
rocks.  Fossil  shells  are  said  to  occur  in  places ;  unfor- 
tunately nothing  was  met  with  except  some  fragments 
insufficient  for  determination. 

At  the  south  end  of  Anncsley  Bay,  just  south  of  the 
Turkish  port  at  Arafile,  there  is  a  cone  composed  of  fine 
volcanic  ash.  From  its  form  this  cone  is  evidently  of 
recent  origin.  About  two  days'  journey  to  the  south 
there  is  said  to  be  an  active  volcano. 

Although  there  can  be  but  little  doubt  of  the  recent 
date  of  a  large  portion  of  this  series,  the  great  amount  of 
denudation  which  parts  of  it  have  undergone  around 
Zulla  and  Massowa,^  where  no  trace  of  the  original 
volcanic  vents  can  be  detected,  and  the  very  considerable 
disturbance  which  has  afiected  the  beds,  prove  that  some 
portions  of  the  series  are  of  ancient  date,  and  that 
the  geological  period  during  which  they  have  accumu- 
lated must  have  been  of  long  duration.  It  is  unfortu- 
uate  that  no  good  specimens  of  the  fossil  shells  occur- 

^  The  same  is  the  case  at  Aden. 


192  OEOLOOr. 

ring  were  found  near  Massowa,  as  by  comparison  of 
them  with  the  species  living  in  the  Red  Sea  some 
idea  could  be  formed  of  the  change  which  has  taken 
place.  The  beds  met  with  near  Massowa  must  have 
been,  in  part  at  least,  of  littoral  or  sub-littoral  origin,  for 
the  shells  which  occur  in  them  are  evidently  marine. 
Yet  if  the  accumulation  took  place  along  the  coast,  it  is 
starange,  imless  the  character  of  that  coast  has  altered 
greatly,  that  shells  and  corals  are  not  exceedingly  abun- 
dant instead  of  being  rare.  It  is  very  probable  that,  while 
some  of  the  sedimentary  beds  are  littoral,  othera  are  of 
sub-aarial  origin,  and  that  their  materials  were  washed 
by  streams  and  floods  over  the  surface  of  the  volcanic 
beds,  like  the  gravels  mentioned  in  the  next  section. 

The  volcanic  series  appears  to  rest  on  metamorphic 
rocks,  in  places  at  least,  both  on  the  Abyssinian  and 
Arabian  coasts.^  In  both  cases,  so  far  as  is  known,  it 
forms  a  fringe  of  varying  breadth. 

Hot  Springs. — The  great  prevalence  of  hot  springs  along 
the  Abyssinian  coast  is  doubtless  connected  with  the  vol- 
canic formation.  Two  of  the  best  known  of  these  are  that 
at  Ailat,  about  thirty  miles  west  of  Massowa  (temp.  150^ 
Fahr.),  and  that  of  Atzfut,  about  eight  miles  south  of  Zulla 
(temp.  140**  Fahr.).  The  latter  is  brackish,  and  probably 
contains  some  sea-water,  as  it  issues  close  by  the  shore ; 
the  former  is  excellent,  and  supplies  the  \dllage  of  Ailat. 

1  Although  the  hills  north  of  Lahej,  twenty-five  miles  from  Aden,  are 
volcanic,  pebbles  of  metamorphics  are  washed  down  by  the  streams  from 
beyond  in  abundance :  so  it  is  evident  that  the  rocks  crop  out  a  little  further 
inland. 


ADEN  SERIES  OF  VOLCANIC  ROCKS,  193 

But  all  subterranean  waters  in  the  neighbourhood  of 
Zulla  appear  to  have  an  unusually  high  temperature. 
The  water  in  some  of  the  wells  at  Komayli  was  decidedly 
warmer  than  that  in  others  when  they  were  first  dug ; 
and  in  a  well  sunk  by  the  Punjab  Pioneers  about  five 
miles  from  the  sea,  in  the  plain  near  the  line  of  railway 
made  by  the  army  from  Zulla  to  Komayli,  the  tempera- 
ture was  about  105V  considerably  higher  than  the 
average  in  air.  The  abundance  of  hot  springs  in  the 
Danakil  country  has  been  noticed  by  M.  Rocher  d'H^ri- 
court  and  others. 


^  I  have  omitted  to  keep  a  record  of  the  exact  temperature,  which  waa 
determined  by  Dr.  Cook. 


194  QEOLOQY. 


SECTION  VIII. 


RECENT  FORMATIONS. 


Albivial  Deposits  near  the  Coast — ^The  plain  around 
Zulla  consists  of  aLluvium,  doubtless  deposited  by  the 
Haddas  and  Komayli  streams.  It  extends  to  the  Gadam 
hills  and  to  the  spur  of  volcanic  rocks  stretching  away 
from  the  south  end  of  the  range.  Beyond  this  spur 
another  plain,  which  is  continuous  with  that  along  the 
seashore  to  the  southward,  extends  to  the  foot  of  the 
mountains  at  Wia  Hadoda  and  KomayU.  This  plain 
consists  of  beds  of  sand  and  gravel,  evidently  deposited 
by  streams  from  the  hills.  At  the  entrance  to  each 
ravine  the  plain  is  raised  considerably  by  the  accumu- 
lation of  detritus  brought  down  by  the  torrents. 

The  spurs  of  the  Gadam  range  almost  reach  the  sea, 
an  exceedingly  narrow  plain  intervening.  This  plain  is 
almost  equally  narrow  near  Arkiko.  The  island  of  Mas- 
sowa  and  the  coast,  for  some  distance  at  least  to  the 
northward,  are  of  coral  formation.  Northwards  firom 
Massowa,  however,  the  country  for  many  miles  from 
the  sea  is  covered  by  an  accumulation  of  pebbles  and 
boulders  of  metamorphic  rocks  and  trap.  In  places,  as 
n^fjlp^pba,  about  thirty  miles  north-north-west  of  Mas- 
beds  form  flat-topped  rises  of  no  great  height, 


RECENT  FORMATIONS.  195 

evidently  the  remains  of  a  plain,  the  greater  portion  of 
which  has  been  cut  away  by  streams  and  surface-waters. 
Further  north  the  plain  still  exists  imbroken  by  ravines, 
and  forms  the  desert  tract  known  as  Shob.  From  the 
quantity  of  boulders  and  pebbles,  and  the  scarcity  of 
corals  and  shells,  there  can  be  but  little  doubt  that  this 
plain  has  not  been  formed  by  the  sea,  but  is  a  deposit 
from  the  wash  of  rain  and  floods  from  the  hills.  As 
already  remarked,  the  sedimentary  beds  interstratified 
with  the  volcanic  rocks  of  the  neighbourhood  are  pro- 
bably, in  great  part  at  least,  of  similar  origin. 

Coral  Formation  of  the  Red  Sea  Coast — ^The  island 
of  Massowa  and  the  coast  in  the  neighbourhood  are 
composed  of  fragments  of  corals,  with  shells  and  sand, 
agglutinated  together  into  an  impure  limestone  rock.  A 
large  number  of  the  islands  fringing  the  coast  in  iJiis 
part  of  the  Red  Sea  are  of  the  same  formation,  and  from 
the  descriptions  given  it  is  highly  probable  that  the 
islands  of  the  Dahalac  Archipelago  are  all  similar. 

All  these  islands  are  perfectly  flat,  and  stand  about 
twenty  or  thirty  feet  above  high-water  mark,^  being  sur- 
rounded by  a  low  cliflf  of  that  height.  A  reef  extends 
for  a  varying  distance  outside  of  them,  and  terminates 
abruptly.  Doubtless  the  island  was  formerly  co-exten- 
sive with  the  reef,  but  so  much  of  the  rock  as  is  exposed 
to  the  influence  of  the  breakers  has  been  swept  away  by 
the  sea.  The  reef  is  naturally  broadest  in  the  direction 
from  which  the  heaviest  breakers  come,  the  open  sea. 

*  In  the  island  of  Massowa,  the  eastern  or  seaward  side  is  much  higher 
than  the  other. 

0  2 


196  HEOLOOr. 

From  the  small  opportunity  that  was  afforded  of  exa- 
mining this  formation,  nothing  can  be  added  to  what  is 
already  known  of  it. 

SoiU  of  the  Abyssinian  Highlands. — The  great  fer- 
tility of  Central  and  Southern  Abyssinia,  the  home,  from' 
time  immemorial,  of  one  of  the^  few  African  nations 
which  have  risen  above  the  savage  state,  is  doubtless 
due,  in  a  great  measure,  to  the  prevalence  of  trap  rocks. 
Upon  the  sandstones  and  limestones  of  Tigr^  vegetation 
is  poor  and  sparse ;  on  the  metamorphic  rocks  the  soil 
is  better ;  but  it  is  only  on  the  traps,  or  in  their  neigh- 
bourhood, that  it  is  really  rich. 

The  most  interesting  circumstance  connected  with  the 
Abyssinian  soils  is,  that  wherever  basaltic  rocks  prevail 
there  is  found  a  very  fertile,  somewhat  argillaceous  soil, 
black  or  dai'k  brown  in  colour,  opening  into  cracks  in 
hot  weather,  and  in  every  way  resembling  the  well- 
known  "regur"  of  Central,  Western,  and  Southern 
India.  Precisely  as  in  India,  moreover,  this  soil  appears 
peculiarly  suited  to  the  growth  of  grasses  and  cereals,^ 
but  less  adapted  for  forest,  perhaps  because  the  luxuriant 
grass  chokes  the  growth  of  the  larger  plants.  This  soil 
does  not  appear  to  be  met  with  far  from  basaltic  rocks, 
although  plains  of  it  may  occur  upon  other  formations, 
as  at  Buya,  the  site  of  the  British  camp  near  Antalo.  In 
this  and  similar  cases,  it  is  highly  probable  that  the  soil 
has  been  derived  from  the  traps  which  cap  the  neigh- 

^  Wherever  the  climate  is  sufficiently  warm,  this  soil  is  doubtless  admirably 
suited  for  cotton  also.  So  much  is  this  the  case  in  India,  that  the  soil  in 
question  is  commonly  known  as  ^'  cotton  soil." 


RECENT  FORMATIONS.  197 

bouriiig  LilLs,  and  probably  formerly  existed  over  the 
l^lain,  but  have  been  removed  by  denudation.  As  no 
thickness  of  soil  is  generally  found  upon  the  limestones, 
that  which  covers  the  Buya  plain  can  scarcely  be  derived 
from  them. 

This  black  soil  of  Abyssinia,  moreover,  appears  to  be 
only  derived  from  the  disintegration  of  dolerites,  not 
from  trachytes.  The  two  are  usually  intermingled,  so  that 
the  soil  upon  the  hills  is  derived  from  both :  but  on  the 
plateau  of  Dalanta  and  Wadela,  between  the  Takkazzyd 
and  Jitta  rivers,  the  route  from  Santara  to  Gaso  traverses 
a  region  of  trachytic  rocks  alone,  and  the  soil  is  light- 
coloured  ;  whilst  from  a  little  south-west  of  Gaso  to  the 
Jitta,  and  again  on  the  plateau  between  the  Jitta  and 
Bashilo,  the  rock  is  basaltic  and  the  soil  black.  Of  course 
it  will  be  found  that  black  soils  are  met  with  elsewhere ; 
any  soil  which  accumulates  under  marshy  conditions  is 
usually  dark-coloured  :  but  the  interesting*  circumstance 
is  the  constant  association  of  a  peculiar  soil,  dark  in 
colour,  and  well  suited  for  the  cultivation  of  cereals  and 
cotton,  with  basalts  and  other  doleritic  rocks. 

Flukes  of  Obsidian. — In  many  places,  small  chips  of 
oljsidian  are  found  scattered  al)out,  frequently  far  from 
any  locality  where  the  rock  is  met  with  in  situ.  From 
their  peculiai-  fonn,  and  the  nature  of  the  facets,  there 
can  l)c  little  or  no  hesitation  in  attributing  these  to 
human  manufacture.  They  are  evidently  the  chijis 
struck  off  in  the  process  of  manufacturing  stone  imple- 
ments, and  are  perfectly  identical  in  shape  with  similar 
chips  found  extensively  in  Europe  and  India. 


198  GEOLOGY. 

A  few  were  met  with  near  ZuUa,  some  were  picked  up 
on  the  highlands,  and  two  or  three  in  the  neighbourhood 
of  Magdala.  But  a  much  larger  number  were  found  at 
Rairo,  near  Af  Abed  in  the  Habab,  in  the  centre  of  a 
granitoid  countr}'-,  and  with  no  volcanic  formation  nearer 
than  the  hills  between  Ain  and  the  sea,  at  least  twenty 
miles  distant.  The  fragments  found  are  of  no  special 
beauty ;  no  well-formed  implements  were  obtained ;  and 
the  occurrence  of  such  chips  is  simply  interesting,  as 
adding  one  more  to  the  numerous  countries  in  which 
traces  of  the  early  use  of  rude  stone  implements  by 
mankind  have  hitherto  been  found. 


190  GEOLOGY. 


SECTION   VII. 

ADEN  SERIES  OF  VOLCANIC  ROCKS 

Along  both  shores  of  the  Red  Sea  and  of  the  Gulf  of 
Aden  there  is  known  to  be  a  great  development  of  vol- 
canic rocka  There  is  every  probability  that  these,  so 
far  as  they  are  known,  belong  to  one  series  ;  and  as  some 
active  volcanoes  still  exist  in  the  sea,  and  cones  quite 
unchanged  in  form,  and  evidently  of  very  recent  date, 
abound  in  many  places  along  the  coast,  it  is  clear  that 
the  series  is  still  in  process  of  formation,  and  that  it  is, 
in  part  at  least,  of  recent  date. 

These  volcanic  formations  are  but  sparingly  represented 
on  the  west  side  of  Annesley  Bay.  They  appear,  how- 
ever, to  compose  the  greater  portion,  if  not  the  whole,  of 
the  Buri  peninsula  east  of  the  bay,  and  they  are  deve- 
loped to  a  considerable  extent  west  of  Massowa. 

In  mineral  character  they  differ  entirely  from  all  the 
volcanic  rocks  of  the  highlands,  and  there  cannot  be  the 
slightest  hesitation  in  assigning  to  them  a  totally  dif- 
ferent and  much  more  recent  geological  age.  They  con- 
.sist^  in  the  neighbourhood  of  Annesley  Bay  and  Massowa, 
of  basaltic  trap  and  of  ash,  both  frequently  scoriaceous, 


ADEN  SERIES  OF  VOLCANIC  ROCKS.  191 

interstratified  with  which  are  sedimentary  rocks,  sand- 
stones, and  conglomerates. 

As  a  rule  these  beds  are  inclined,  and  present  the 
appearance  of  having  been  much  disturbed.  The  beds 
of  the  Buri  peninsula  were  only  examined  in  one  spot. 
Those  of  the  neighbourhood  of  Zulla  are  most  irregular 
in  dip,  whilst  in  the  tract  west  and  north-west  of  Mas- 
sowa,  with  some  irregularity,  there  is  a  very  constant 
general  dip  towards  the  sea  to  the  eastward. 

The  sedimentary  beds  contain  a  mixture  of  sand,  ashes, 
and  rolled  fragments  of  both  volcanic  and  metamorphic 
rocks.  Fossil  shells  arc  said  to  occur  in  places ;  unfor- 
tunately nothing  was  met  with  except  some  fragments 
insufficient  for  determination. 

At  the  south  end  of  Annesley  Bay,  just  south  of  the 
Turkish  port  at  Arafil^,  there  is  a  cone  composed  of  fine 
volcanic  ash.  From  its  form  this  cone  is  evidently  of 
recent  origin.  About  two  days'  journey  to  the  south 
there  is  said  to  be  an  active  volcano. 

Although  there  can  be  but  little  doubt  of  the  recent 
date  of  a  large  portion  of  this  series,  the  great  amount  of 
denudation  which  parts  of  it  have  undergone  around 
Zulla  and  Massowa,^  where  no  trace  of  the  original 
volcanic  vents  can  be  detected,  and  the  very  considerable 
disturbance  which  has  afiected  the  beds,  prove  that  some 
portions  of  the  series  are  of  ancient  date,  and  that 
the  geological  period  during  which  they  have  accumu- 
lated must  have  been  of  long  duration.  It  is  unfortu- 
Date  that  no  good  specimens  of  the  fossil  shells  occur- 

^  The  same  is  the  case  at  Aden. 


192  OEOLOOr, 

ring  were  found  near  Massowa,  as  by  comparison  of 
them  with  the  species  living  in  the  Red  Sea  some 
idea  could  be  formed  of  the  change  which  has  taken 
place.  The  beds  met  with  near  Massowa  must  have 
been,  in  part  at  least,  of  littoral  or  sub-littoral  origin,  for 
the  shells  which  occur  in  them  are  evidently  marine. 
Yet  if  the  accumulation  took  place  along  the  coast,  it  is 
strange,  imless  the  charaxjter  of  that  coast  has  altered 
greatly,  that  shells  and  corals  are  not  exceedingly  abun- 
dant instead  of  being  rare.  It  is  very  probable  that,  while 
some  of  the  sedimentary  beds  are  littoral,  others  are  of 
sub-aerial  origin,  and  that  their  materials  were  washed 
by  streams  and  floods  over  the  surface  of  the  volcanic 
beds,  like  the  gravels  mentioned  in  the  next  section. 

The  volcanic  series  appears  to  rest  on  metamorphic 
rocks,  in  places  at  least,  both  on  the  Abyssinian  and 
Arabian  coasts.^  In  both  cases,  so  far  as  is  known,  it 
forms  a  fringe  of  varying  breadtL 

HotSprings. — The  great  prevalence  of  hot  springs  along 
the  Abyssinian  coast  is  doubtless  connected  with  the  vol- 
canic formation.  Two  of  the  best  known  of  these  are  that 
at  Ailat,  about  thirty  miles  west  of  Massowa  (temp.  150^ 
Fahr.),  and  that  of  Atzfut,  about  eight  miles  south  of  Zulla 
(temp.  140**  Fahr.).  The  latter  is  brackish,  and  probably 
contains  some  sea-water,  as  it  issues  close  by  the  shore ; 
the  former  is  excellent,  and  supplies  the  village  of  Ailat. 

^  Although  the  hills  north  of  Lahej,  twenty-five  miles  from  Aden,  are 
volcanic,  pebbles  of  metamorphics  are  washed  down  by  the  streams  from 
beyond  in  abundance :  so  it  is  evident  that  the  rocks  crop  out  a  little  further 
inland. 


ADEN  SERIES  OF  VOLCANIC  ROCKS,  193 

But  all  subterranean  waters  in  the  neighbourhood  of 
Zulla  appear  to  have  an  unusually  high  temperature. 
The  water  in  some  of  the  wells  at  Komayli  was  decidedly 
warmer  than  that  in  others  when  they  were  first  dug ; 
and  in  a  well  sunk  by  the  Punjab  Pioneers  about  five 
miles  from  the  sea,  in  the  plain  near  the  line  of  railway 
made  by  the  army  from  Zulla  to  Komayli,  the  tempera- 
ture was  about  105V  considerably  higher  than  the 
average  in  air.  The  abundance  of  hot  springs  in  the 
Danakil  country  has  been  noticed  by  M.  Rocher  d'Hdri- 
court  and  others. 


^  I  have  omitted  to  keep  a  record  of  the  exact  temperature,  which  waa 
determined  by  Dr.  Cook. 


194  QEOLOQY. 


SECTION  VIII. 


RECENT  FORMATIONS. 


Alluvial  Deposits  near  the  Coast — ^The  plain  around 
Zulla  consists  of  alluvium,  doubtless  deposited  by  the 
Haddas  and  Komayli  streams.  It  extends  to  the  Gadam 
hills  and  to  the  spur  of  volcanic  rocks  stretching  away 
fix)m  the  south  end  of  the  range.  Beyond  this  spur 
another  plain,  which  is  continuous  with  that  along  the 
seashore  to  the  southward,  extends  to  the  foot  of  the 
mountains  at  Wia  Hadoda  and  Komayli.  This  plain 
consists  of  beds  of  sand  and  gravel,  evidently  deposited 
by  streams  from  the  hills.  At  the  entrance  to  each 
ravine  the  plain  is  raised  considerably  by  the  accumu- 
lation of  detritus  brought  down  by  the  torrents. 

The  spurs  of  the  Gadam  range  almost  reach  the  sea, 
an  exceedingly  narrow  plain  intervening.  This  plain  is 
almost  equally  narrow  near  Arkiko.  The  island  of  Mas- 
sowa  and  the  coast,  for  some  distance  at  least  to  the 
northward,  are  of  coral  formation.  Northwards  from 
Massowa,  however,  the  country  for  many  miles  from 
the  sea  is  covered  by  an  accumulation  of  pebbles  and 
boulders  of  metamorphic  rocks  and  trap.  In  places,  as 
near  Amba,  about  thirty  miles  north-north-west  of  Mas- 
sowa, these  beds  form  flat-topped  rises  of  no  great  height, 


RECENT  FORMATIONS.  195 

evidently  the  remains  of  a  plain,  the  greater  portion  of 
which  has  been  cut  away  by  streams  and  surface-waters. 
Further  north  the  plain  still  exists  unbroken  by  ravines, 
and  forms  the  desert  tract  known  as  Shob.  From  the 
quantity  of  boulders  and  pebbles,  and  the  scarcity  of 
corals  and  shells,  there  can  be  but  little  doubt  that  this 
plain  has  not  been  formed  by  the  sea,  but  is  a  deposit 
from  the  wash  of  rain  and  floods  from  the  hills.  As 
already  remarked,  the  sedimentary  beds  interstratified 
with  the  volcanic  rocks  of  the  neighbourhood  are  pro- 
bably, in  great  part  at  least,  of  similar  origin. 

Coral  Formation  of  the  Red  Sea  Coast. — The  island 
of  Massowa  and  the  coast  in  the  neighbourhood  are 
composed  of  fragments  of  corals,  with  shells  and  sand, 
agglutinated  together  into  an  impure  limestone  rock.  A 
large  number  of  the  islands  fringing  the  coast  in  this 
part  of  the  Ked  Sea  are  of  the  same  formation,  and  from 
the  descriptions  given  it  is  highly  probable  that  the 
islands  of  the  Dahalac  Archipelago  are  all  similar. 

All  these  islands  are  perfectly  flat,  and  stand  about 
twenty  or  thirty  feet  above  high-water  mark,^  being  sur- 
rounded by  a  low  cliff  of  that  height.  A  reef  extends 
for  a  varying  distance  outside  of  them,  and  terminates 
abruptly.  Doubtless  the  island  was  formerly  co-exten- 
sive with  the  reef,  but  so  much  of  the  rock  as  is  exposed 
to  the  influence  of  the  breakers  has  been  swept  away  by 
the  sea.  The  reef  is  naturally  broadest  in  the  direction 
from  which  the  heaviest  breakers  come,  the  open  sea. 

*  In  the  island  of  Massowa,  the  eastern  or  seawaxd  side  is  much  higher 
than  the  other. 

0  2 


196  GEOLOGY, 

From  the  small  opportimity  that  was  afforded  of  exa- 
mining this  formation,  nothing  can  be  added  to  what  is 
already  known  of  it. 

Soils  of  the  Abi/ssi7iian  Highlands. — The  great  fer- 
tility of  Central  and  Southern  Abyssinia,  the  home,  from' 
time  immemorial,  of  one  of  the^  few  African  nations 
which  have  risen  above  the  savage  state,  is  doubtless 
due,  in  a  great  measure,  to  the  prevalence  of  trap  rocks. 
Upon  the  sandstones  and  limestones  of  Tigr^  vegetation 
is  poor  and  sparse ;  on  the  metamorphic  rocks  the  soil 
is  better ;  but  it  is  only  on  the  traps,  or  in  their  neigh- 
bourhood, that  it  is  really  rich. 

The  most  interesting  circumstance  connected  with  the 
Abyssinian  soils  is,  that  wherever  basaltic  rocks  prevail 
there  is  found  a  very  fertile,  somewhat  argillaceous  soil, 
black  or  dai-k  brown  in  colour,  opening  into  cracks  in 
hot  weather,  and  in  every  way  resembling  the  well- 
known  "regiu*''  of  Central,  Western,  and  Southern 
India.  Precisely  as  in  India,  moreover,  this  soil  appears 
peculiarly  suited  to  the  growth  of  grasses  and  cereals,^ 
but  less  adapted  for  forest,  perhaps  because  the  luxuriant 
grass  chokes  the  growth  of  the  larger  plants.  This  soil 
does  not  appear  to  be  met  with  far  from  basaltic  rocks, 
although  plains  of  it  may  occur  upon  other  formations, 
as  at  Buya,  the  site  of  the  British  camp  near  Antalo.  In 
this  and  similar  cases,  it  is  highly  probable  that  the  soil 
has  been  derived  from  the  traps  which  cap  the  neigh- 

^  Wherever  the  climate  is  sufficiently  warm,  this  soil  is  doubtless  admirably 
suited  for  cotton  also.  So  much  is  this  the  case  in  India,  that  the  soil  in 
question  is  commonly  known  as  **  cotton  soil." 


RECENT  FORMATIONS,  197 

bouring  hills,  and  probably  formerly  existed  over  the 
l^lain,  but  have  been  removed  by  denudation.  As  no 
thickness  of  soil  is  generally  found  upon  the  limestones, 
that  which  covers  the  Buya  plain  can  scarcely  be  derived 
from  them. 

This  black  soil  of  Abyssinia,  moreover,  appears  to  be 
only  derived  from  the  disintegration  of  dolerites,  not 
from  trachytes.  The  two  are  usually  intermingled,  so  that 
the  soil  upon  the  hills  is  derived  from  both :  but  on  the 
plateau  of  Dalanta  and  Wadela,  between  the  Takkazzyd 
and  Jitta  rivers,  the  route  from  Santara  to  Gaso  traverses 
a  region  of  trachytic  rocks  alone,  and  the  soil  is  light- 
coloured  ;  whilst  from  a  little  south-west  of  Gaso  to  the 
Jitta,  and  again  on  the  plateau  between  the  Jitta  and 
Bashilo,  the  rock  is  basaltic  and  the  soil  black.  Of  course 
it  will  be  found  that  black  soils  are  met  w^ith  elsewhere  ; 
any  soil  which  accumulates  under  marshy  conditions  is 
usually  dark-coloured  :  but  the  interesting*  circumstance 
is  the  constant  association  of  a  peculiar  soil,  dark  in 
colour,  and  well  suited  for  the  cultivation  of  cereals  and 
cotton,  with  basalts  and  other  doleritic  rocks. 

Flakes  of  Obsidian. — In  many  places,  small  chips  of 
obsidian  are  found  scattered  about,  frequently  far  from 
any  locality  where  the  rock  is  met  with  in  situ.  From 
their  peculiar  form,  and  the  nature  of  the  facets,  there 
can  be  little  or  no  hesitation  in  attributing  these  to 
human  manufacture.  They  are  evidently  the  chips 
struck  off  in  the  process  of  manufacturing  stone  imple- 
ments, and  are  perfectly  identical  in  shape  with  similar 
chips  found  extensively  in  Europe  and  India. 


198  GEOLOGY. 

A  few  were  met  with  near  Zulla,  some  were  picked  up 
on  tlie  highlands,  and  two  or  three  in  the  neighbourhood 
of  Magdala.  But  a  much  larger  number  were  found  at 
Bairo,  near  Af  Abed  in  the  Habab,  in  the  centre  of  a 
granitoid  country,  and  with  no  volcanic  formation  nearer 
than  the  hills  between  Ain  and  the  sea,  at  least  twent}' 
miles  distant  The  fragments  found  are  of  no  special 
beauty ;  no  well-formed  implements  were  obtained ;  and 
the  occurrence  of  such  chips  is  simply  interesting,  as 
adding  one  more  to  the  numerous  countries  in  which 
traces  of  the  early  use  of  rude  stone  implements  by 
mankind  have  hitherto  been  found. 


APPENDIX. 


DESCRIPTIONS  OF  THE  NEW  SPECIES  OF  FOSSILS 
FROM  THE  ANTALO  LIMESTONE. 


Hemicidaris  Abyssinica,  sp.  nov.,  fig.  1. 

Testa  depresso-glohom,  vtl  sphceroidea,  vel  suhttu  plana,  mprafert  hemi* 
8j)/icerica.  Arece  amhulacraUs  suhrectce^  strictas,  tubercults  parvis, 
mbcequalibusy  vix  mptme  minor ibuSy  fire  cequidistantibus  biseriatim 
altemantibus  et  granulU  irregulaiiter  interspertU  omaias.  Area 
interamhulacrales  seriehus  duobus  8-9  tubercularum  major um  in- 
structAje  tuberadia  utrinque  granulis  circumdatis,  aliis  granulis 
minoribus  scepe  irregulariter  ad  margines  laminarum  interjectu  ; 
laminis  latioribus  quam  altis,  Oa  magnum  decies  incimm,  Spinai 
elongatce,  ct/Hndricce,  longiiudinaliter  subobsolete  ttriatce,  circa 
1^  mm.  a  basi  angvlatim  dilatatce,  ibidem  et  ad  badn  granulii 
circumdatce  ;  spatio  iyitermedio  loevi,  lente  et  regtdariter  attemuUo, 

Var.  depresses  ratio  diam.  :  alt.  : :  1   :  1*9. 
Var,  sub/iemisphcericce  id.  :  id.   ::  I  :  I'b. 

The  principal  characters  of  this  form  are  the  small  and  generally 
equal  size  of  the  ambulacral  tubercles,  and  their  arrangement: 
those  in  the  two  rows  alternating  at  subequal  distances,  to- 
wards the  lower  portion  of  the  ambulacral  areas  especially ;  near 


200  GEOLOGY, 

the  anus  tbey  become  much  smaller,  and  run  much  closer 
together.  The  rows  of  ambulacral  pores  are  not  flexuous  above, 
but  close  to  the  mouth  they  become  zigzagged  and  slightly 
irregular.  The  interambulacral  plates  are  broader  than  high. 
The  number  of  tubercles,  which  varies  from  eight  to  nine,  or 
even  ten,  including  a  very  small  one  close  to  the  mouth  (being 
rarely  seven  in  very  small  specimens),  is  greater  than  in  most 
allied  forms.  The  spines,  many  of  which  occur  with  the  shells, 
are  long,  with  the  remains  of  longitudinal  striation  visible  upon 
them  :  the  shoulder  near  the  base  is  rather  wide  and  granulated, 
the  base  also  granulated,  the  intermediate  portion  shaped  like 
Die  frustum  of  a  cone.  The  granulation  of  the  shoulder  is  very 
slight,  and  easily  disappears  with  weathering,  so  that  many 
spines  appear  as  if  not  granulated  hei*e:  the  granulation  cor- 
I'esponds  to  the  terminations  of  the  longitudinal  stride.  The 
apical  disc  is  ill-preserved  or  wanting  in  all  the  specimens 
found. 

The  nearest  approach  to  this  form  appears  to  me  to  be  made 
by  H,  Wrifjhtii,  Cotteau  (not  of  Desor),  "  Echin.  foss.  Dep.  Yonne," 
p.  294,  pi.  xlii. ;  the  only  diflferences  being  that  that  species 
has  broader,  and  consequently  fewer,  interambulacral  plates, 
that  the  rows  of  pores  bordering  the  ambulacral  areas  are  much 
more  flexuous,  that  the  ambulacral  tubercles  do  not  alteniate, 
and  that  the  spines  are  differently  terminated  at  the  base.  From 
most  other  species,  such  as  H,  Luciensis,  D'Orb.,  H,  intervicdia, 
Forbes,  &c.,  the  present  form  is  distinguished  both  by  its  more 
numerous  interambulacral  plates,  and  also  at  once  by  the  small 
and  subequal  size  of  the  ambulacral  tubercles,  as  well  as  by  the 
form  of  the  spines  where  the  latter  are  known. 

Hcmicidaris  Ahyss^inica — found  rather  abundantly  in  a  bed 
at  the  base  of  the  Antalo  limestone,  close  to  the  camp  at  Mai 
Dongolo,  four  marches  north  of  Antalo  and  three  south  of 
Adigrat.  Many  of  the  specimens  are  in  excellent  order,  and 
tlie  spines  abounded  in  the  bed,  in  some  cases  attached.  Spines, 
of  the  same  species  apparently,  were  met  with  here  and  there 
throughout  the  Antalo  limestone. 


APPENDIX.  201 


EXPLANATION   OF  FIGURES. 

a.  Subhemispheiical  rarietj,  natural  size. 

6.  Depressed  spheroidal  ditto,      ditto. 

c.  View  from  below  ditto. 

d  Ambulacral  segment   and  four  interambulocral   plates,  enlarged   two 

diameters. 
e.  Apical  disk,  enlarged  two  diameters. 
/.  Spine,  natural  size. 
f7.  Base  of  ditto,  enlarged  two  diameters. 


MoDiOLA  IMBRICARIA,  Sow.  var.,  fig.  2. 

This  is  so  near  some  English  specimens  that  there  appears  no 
reason  for  distinguishing  it.  The  rib  running  from  the  umbo  to 
the  neutral  margin  is  rather  nearer  to  the  posterior  side,  and  the 
shell  is  generally  broader  in  proportion  at  the  extremity  remote 
from  the  beaks.  I  consequently  give  a  figure,  but  I  think  it 
unnecessary  to  propose  a  new  name. 


Mytilus  (?  Modiola)  Tigrensis,  sp.  nov.,  fig.  3. 

Tata  irregulariitr  'pyriformUy  conico-^ubovatay  concentrice plicalo  slriataf 
versus  umbones  tumida^  antice  convexOy  postice  plana  ohUqua^  margine 
antico  longe  recto,  tunc  convexo  postico  fere  recto,  post  medium  oon- 
caiiusculOj  ventrali  regulariter  convexo. 

Long.  50,  diam,  postic.  26,  crass.  27  milUm. 

A  few  well-preserved  specimens  were  found  with  the  last. 
This  shell  approaches  M.  subkcms,  Sow.,  but  it  is  more  rounded, 
the  posterior  side  not  being  so  flat,  and  the  umbones  are  more 
obtuse. 


202  OEOLOQY. 


Pholadomya  granulifeba,  sp.  UOV.,  fig.  4 

Tuta  incequilaUraliSf  parum  tumida,  ovato-elongata,  concentrice  ttricUa, 
punctii  in  aeriebus  confertU  radiantibru  dispositis  pulchre  omata, 
costd  obttud  und  submediand  ab  umbone  ad  marginem  decwrrtnU^ 
alidque  paaterioi'i  imtructa ;  latere  antico  juxta  umbones  hremUr 
Miqtu),  tunc  rotutidato  postico  dongatOy  margine  donali  concavius- 
ado  ventrali  convexo,  extremUaU  posticd  oblique  truncata,  umbones 
prominmteSf  canvergentes ;  area  immeraa  glabra. 

Lang,  40,  kU,  25,  crass,  17  mUlem, 

Found  near  Agula.  Only  two  specimens  were  obtained.  A 
veiy  peculiar  and  xmiisual  form  of  Pholadomya^  with  radiating 
rows  of  puncticnlations,  somewhat  allied  to  the  shell  figured 
by  Buvignier  as  Parwpa^a  pu/nctifera,  Buv.,  from  the  coral  rag 
("  Statistique  GeoL,  &c.,  Dep.  de  la  Meuse,"  pi.  viiL  figs.  8 — 

11). 

Pholadomya  sublieata,  sp.  nov.,  fig.  5. 

Testa  valde  ineequHatercUis,  tnediocriter  tumida,  elongato-suhovata,  con- 
centrice striata,  costis  radiantibus  subohsoleiis  medianis  dttdbus, 
postids  confertioribtLS  3-4  difficilissime  notandis,  anticd  nulld, 
omnibus  versus  marginem  ventralem  evanescentihus ;  latere  antico 
juxta  umbones  dedivi,  tunc  rotundato,  dorsali  subhorizontali,  ex- 
trtmitate  posticd  subangvlatd,  margine  ventrali  convexo ;  umbones 
parum  prominentes,  convexi. 

Long,  33,  laL  25,  crass,  19  millem^ 

Found  with  the  last.  Although  much  less  peculiar  than 
P.  grantUifera,  this  is  also  a  well-marked  form,  distinguished  by 
the  few  and  subobsolete  ribs,  thus  being  one  of  the  passage 
forms  between  Pholadomya  and  Myacitcs,  The  nearest  described 
species  appears  to  be  Ph,  cmicatenata,  Agass.,  from  the  lower 
oolite  of  Poland. 


APPENDIX,  203 


CeKOMYA    PAUCILIKATA,   Sp.   IIOV.,  fig.   C. 

C.  excentricse  perdmilu,  sed  liiis  concentricis  mag^U  distantilnis,  luiud 
postice  angulatU,  Testa  mbrJiomboidaHs,  trunaato-ovatay  valde  in- 
cequilateralis,  umhonibus  tumidis,  mhterminalilAis,  margine  dorsali 
honz(yntali,  postico  i^tundato^'  ventrali  jxinim  com^^o,  antico  fere 
vtrticali. 

Long,  75,  lat.  53,  crasf.  circa  60  miUfin,' 

A  single  specimen  was  obtained  with  the  PholadmnycCy  and 
accompanied  also  by  Ceromya  siinilis  and  C.  c(mr4mti*ica. 


PART  III. 


ZOOLOGY. 


ZOOLOGY. 


INTRODUCTION. 

The  earliest  contribution  to  the  Abyssinian  fauna  was 
contained  in  the  last  (fifth)  volume  of  Brace's  Travels. 
In  the  original  edition  of  1790  several  plates  repre- 
senting various  plants  and  animals  of  North-eastern 
Africa  were  given,  accompanied  by  descriptions,  which, 
like  all  of  Brace's  writings,  show  great  power  of  obser- 
vation, though  they  are  occasionally  not  quite  accurate 
in  matters  of  detail.  A  few  of  the  drawings  of  birds 
brought  back  by  Brace  were  described  by  Buffon  and 
named  by  Gmelin,  but  the  exact  species  have  in  some 
instances  remained  obscure. 

In  1814  an  additional  instalment  to  our  knowledge 
of  the  avi-fauna  was  furnished  in  the  appendix  to  Salt's 
"  Travels  in  Abyssinia,''  in  which  an  enumeration  of  the 
birds  coUected  was  given,  and  a  few  of  the  new  forms 
were  described  by  Lord  Stanley. 

After  a  beginning  had  thus  been  made  by  English 
travellers,  the  field  appears  to  have  been  abandoned  by 


208  ZOOLOGY, 

them,  and  tlie  whole  labour  of  collecting  and  describing 
the  fauna  of  Abyssinia  has,  almost  without  exception, 
been  performed  by  Gennan  and  French  travellers  and 
naturalists.  Foremost  amongst  these  stands  RUppell, 
who  made  two  long  journeys  in  North-eastern  Africa. 
The  natural  history  results  of  his  travels  were  published, 
principally  in  three  works,  to  which  frequent  reference 
will  be  made  in  the  following  pages.  These  works 
are — 

1.  "Atlas  zu  der  Reise  in  nordlichen  Africa,  zoo- 
logischer  Theil"  (Atlas  to  the  Journey  in  Northern 
Africa,  Zoological  Portion).     1826. 

2.  "  Neue  Wirbelthiere  zu  der  Fauna  von  Abyssinien 
gehorig''  (New  Vertebrata  belonging  to  the  Abyssinian 
Fauna).  1835.  From  references  made  in  the  text  of 
this  work,  it  is  evident  that  a  portion  was  of  later  date 
than  the  title-page. 

3.  "  Systematische  Uebersicht  der  Vogel  Nord-ost 
Africa's"  (Systematic  Review  of  the  Birds  of  North- 
eastern Africa).  1845.  This  is  a  complete  list  of  all 
the  birds  at  that  time  known  to  have  been  foimd  in  the 
Nile  valley,  Abyssinia,  and  the  countries  on  the  shores  of 
the  Red  Sea.     The  number  enumerated  is  532. 

A  large  number  of  alterations  and  additions  have 
since  been  made  to  this  list,  but  it  contains  a  very 
large  amount  of  valuable  information.  Several  new 
species  were  described  in  each  of  the  three  works,  those 
in  the  " Systematische  Uebersicht"  being  all  birds, 
admirably  figured  by  Wolf;  whilst  in  the  two  earlier 
works  numerous  kinds  of  mammals,  reptile?,  and  fishes 


INTRODUCTION.  209 

were  described  and  figured.  A  few  invertebrata  are 
also  included  in  the  "  Atlas/' 

Besides  these  three  works,  a  few  scattered  papers  have 
been  contributed  by  Riippell  to  the  Journals  of  Societies, 
&c.  Of  these  the  most  important  appeared  in  the 
"Museum  Senckenbergiamun,"  a  scientific  work  pub- 
lished at  Frankfort  (1834-37). 

About  1828-30  appeared  the  "Symbolse  PhysicsB" 
of  Hemprich  and  Ehrenberg,  a  work  of  much  research ; 
but  which,  unfortunately,  was  given  to  the  world  in  a 
very  incomplete  state.  It  contains  elaborate  inscrip- 
tions,  in  Latin,  of  several  mamn^als,  birds,  and  insects, 
collected  by  the  writers  on  the  shores  of  the  Red  Sea, 
where  Professor  Hemprich  died,  one  of  th,e  many 
martyrs  to  science  who  have  fallen  in  Africa.  Sppie, 
both  of  the  mammals  and  birds,  had  akeacjy  been 
described  in  Rtippell's  *'  Atlas."  The  "  SymbolaB 
Physicse"  is  a  very  rare  work,  and  m^y  of  the 
species  of  bird^  described,  only  a  few  of  wl^ch  are 
figured,  have  been  overlooked  by  subsequent  waiters. 

Reference  has  already  been  made  at  t^^e  commence- 
ment of  the  geological  part  of  this  work  to  the  labours 
of  Messrs.  Ferret  and  Galinier.  The  same  officers  made 
large  botanical  and  zoological  collections,  descriptions 
of  which  were  published  in  the  third  volume  of  their 
Travels  in  1847.  The  animajs  \7ere  identified  by 
MM.  Guerin  MeneviUe  and  De  la  F^-esnaye,  and 
descriptions  of  some  new  species  of  birds  were  pub- 
lished by  the  former  in  the  Revue  et  Magasin  de 
Zoologie  for  1843,  pp.  161,  162.     These  were  figured 

p 


210  ZOOLOGY. 

in   the    atlas    which    accompanied    MM.    Ferret    and 
Galinier's  work, 

The  collections  of  Messrs.  Dillon  and  Petit,  attached 
to  Lefebvre's  exploring  expedition,  both  of  whom,  un- 
fortunately, lost  their  lives  in  Abyssinia,  were  examined 
and  determined  by  MM.  Desmurs,  F.  Prevost,  Guichenol, 
and  Guerin  Meneville,  and  an  accoimt  of  them  pub- 
lished in  the  sixth  volume  of  Lefebvre's  "  Voyage  en 
Abyssinie."  The  accompanying  atlas  contained  figures 
of  some  of  the  animals  supposed  to  be  new  to  science. 
It  is  impossible  to  speak  in  high  terms  of  this  work, 
a  large  number  of  the  names  being  inaccurate.  The 
detenninations  aJso  are  too  frequently  based  upon  figures, 
always  a  source  of  error. 

In  1863  Dr.  A.  Brehm  of  Berlin  accompanied  the 
Duke  of  Saxe  Coburg  in  a  shooting  excursion  firom 
Massowa  to  the  highlands  of  Mensa,  east  of  the  Anseba 
valley.  The  journey  was  too  hurried  for  specimens  to 
be  collected,  but  Dr.  Brehm's  previous  extensive  know- 
ledge of  the  North  African  fauna  enabled  him  to 
identify,  for  the  most  part  correctly,  thirty-eight  species 
of  mammals,  and  one  hundred  and  seventy-two  species 
of  birds,  observed  during  his  visit,  and  to  publish, 
in  1863,  in  a  work  called  "Keise  nach  Habesch"\ 
(A  Journey  to  Abyssinia),  a  somewhat  elaborate  ac- 
count of  the  habits  of  the  majority.  In  a  few  instances 
the  descriptions  given  are  based  upon  observations  too 

^  This  name  is  ill  chosen,  becaase  Dr.  Brehm  did  not  enter  Abyssinia 
Proper.  Mensa,  like  Bogos,  is  a  dependency  of  Tigr^,  inhabited  by  a  race  of 
people  distinct  from  the  Abyssinians. 


INTRODUCTION.  211 

hurriedly  made  to  be  quite  trustworthy;  and  it  is 
probable  that  in  some  cases  Dr.  Brehm  did  not  identify 
the  animals  observed  correctly,  an  error  which  want  of 
specimens  prevented  him  from  remedying  subsequently. 

I  have  left  to  the  last  all  notice  of  by  far  the  largest 
and  most  important  contributions  to  our  knowledge  of 
the  Abyssinian  faima  which  have  appeared  since  KUp- 
pell's  time  ;  those  of  Hofrath  Th.  von  Heuglin,  for  many 
years  a  resident  in  the  Nile  valley,  and  who  has  recently 
explored  a  large  portion  of  the  Abyssinian  highlanda 
His  first  important  paper— a  second  "  Systematische 
Uebersicht,"  or  Systematic  Catalogue  of  the  North-eaat 
African  Avi-fauna — appeared  in  the  Journal  der  zoolch 
gischen  hotanischen  Gesellschaft  zu  Wien^  for  1856, 
pp.  255  to  354.  In  this  list  754  species  of  birds  are 
enumerated  as  occurring  in  North-eastern  Africa.  I 
cannot  help  thinking,  however,  that  the  publication  of 
this  paper  is  to  be  regretted.  For  the  writer  there  is 
some  excuse,  as  the  list  was  doubtless  drawn  up  with- 
out sufficient  means  of  reference,  but  the  Society  which 
allowed  it  to  appear  in  its  Journal  deserves  the  repro- 
bation of  every  naturalist  who  does  not  wish  orni- 
thological nomenclature  to  become  a  mass  of  confusion. 
Every  mistake  of  Ruppell's  was  repeated,  and  a  very 
large  number  of  species  were  named  as  new  without 
descriptions  being  furnished.  The  majority  of  these 
supposed  species  have  since  been  foimd,  by  the  author 
himself,  to  have  been  proposed  in  error.  Many  were 
only  varieties,  others  had  already  been  described,  and 

*  Journal  of  the  Zoological  and  Botanical  Society  of  Vienna, 

p2 


212  ZOOLOGY. 

80  little  value  is  atttocbed  to  the  names  proposed  by 
Von  Heuglin  that,  when  furnishing  descriptions  of 
some  of  the  species  several  years  lat<?r,  he  has  himself 
changed  a  great  proportion  of  those  given  by  him  in  his 
former  essay.  In  the  following  page^  I  shall  refer  to 
this  paper  as  little  as  possible. 

Short  notices  by  Von  Heuglin,  with  occasional  de- 
scriptions of  species  both  of  birds  and  mammals,  are 
scattered  through  Cabanis'  "  Journal  fiir  Omithologie," 
the  "  Ibis,"  the  "  Nova  Acta  Academiae  Leopoldo-Caro- 
linee,''  and  "  Petermann's  Mittheilungen,"  during  the  last 
ten  years.  And  recently  (1867-69)  some  most  valuable 
and  exhaustive  descriptions  of  all  the  species  belonging 
to  several  families  of  Insessorial  Birds  inhabiting  North- 
eastern  Africa  by  the  same  naturalist  have  appeared  in 
the  "  Journal  fur  Omithologie,"  and  a  synopsis  of  the 
MalurincB  in  the  "  Ibis"  for  1869.  During  the  course 
of  this  year  also  the  same  writer  has  published  the  first 
eight  parts  of  an  elaborate  work  on  the  birds  of  North- 
eastern Africa,  called  "  Omithologie  Nord-ost  Africa's." 
To  all  of  these  later  papers,  and  to  the  more  extensive 
work  just  mentioned,  I  have  supplied  references  in  the 
following  pages.  I  cannot  in  all  cases  coincide  in  Von 
Heuglin's  nomenclature.  His  descriptions,  however,  are 
excellent,  and  detailed,  and  numerous  references  are 
given  to  the  works  of  previous  writers. 

In  the  subsequent  pages  only  the  vertebrata  of  my 
Abyssinian  collections  have  been  fully  enumerated  and 
properly  worked  out.  Despite  much  time  and  labour, 
the  synonomy  is  far  fi-om  being  so  thorough  as  I  could 


INTRODUCTION.  213 

have  wished,  but  only  those  who  have  entered  into  such 
questions  know  the  amount  of  time  w^hich  must  be  given 
to  each  species  before  its  synonomy  can  be  accurately 
determined.  For  the  determinations  of  many  species  of 
birds  I  am  indebted  to  the  aid  of  Messrs.  Jules  Verreaux, 
G.  R  Gray,  Lord  Walden,  Messrs.  Gumey,  Tristram, 
Harting,  K.  B.  Sharpc,  Cabanis,  and  especially  to  Dr. 
Finsch,  to  whose  MS.  list  and  notes^  of  the  birds  col- 
lected by  my  friend  Mr.  Jesse  I  have  had  access  through 
the  kindneas  of  Dr.  Sclater.  For  the  identification  of 
the  reptiles  and  of  the  two  species  of  fish  brought  back 
by  me,  my  thanks  are  due  to  Dr.  GUnther.  For  aid 
and  information  in  determining  the  moUusca,  I  am 
under  obligations  to  Messrs.  G.  Nevill,  Hanley,  and 
H.  Adams. 

I  mucli  regret  that  I  am  unable  at  present  to  furnish 
a  detailed  list  of  the  invertebrata  collected,  and  that  of  the 
moUusca  at  least,  to  which  I  have  devoted  much  time,  I 
cannot  give  as  full  an  account  as  of  the  vertebrata.  But 
the  leave  which  has  been  most  liberally  granted  to  me 
by  the  Government  of  India  for  the  purpose  of  preparing 
this  work,  together  with  an  extension  sanctioned  by  the 
Secretaiy  of  State  for  India,  being  at  an  end,  I  find 
that  the  time  has  not  sufficed  for  that  study  of  the  very 
extensive  and  confused  literature  of  the  subject  which 
alone  would  enable  me  to  do  it  justice.  I  have  conse- 
quently been  compelled  to  limit  myself  to  the  appendcMl 
list. 

^  Tliis  list  will  appear  in  the  Transactions  of  the  Zoological  Society  of 
London. 


214  ZOOLOGY. 

The  physical  geography  of  Abyssinia  has  already  been 
briefly  described  in  the  commencement  of  the  geological 
portion  of  this  work.  It  is  only  necessary  to  refer  to  it 
here  in  order  to  point  out  that  a  large  portion  of  the 
country  consists  of  plateaux  7,000  to  10,000  feet  above 
the  sea,  and  that  from  amongst  these  a  few  peaks  and 
ranges  rise  to  12,000  and  even  15,000  feet,  whilst  many 
of  the  valleys,  even  where  they  intersect  lofty  plateaux, 
are  cut  to  so  great  a  depth  that  at  the  bottom  they  do 
not  exceed  4,000  or  5,000  feet  above  the  sea.  As  may 
be  readily  supposed,  there  is  a  great  diflFerence  in  the 
animals  and  plants  inhabiting  these  different  elevations, 
and  the  distinction  has  attracted  far  less  attention 
amongst  naturalists  than  it  deserves. 

My  own  observations  were  naturally  imperfect  It 
not  unfrequently  happens  that  an  animal,  which  fix)m 
peculiarities  of  soil  and  climate  is  constant  to  one  eleva- 
tion in  one  region,  ranges  higher  or  lower  in  others ;  and 
this  I  found  to  be  the  case  in  several  instances.  Had  my 
explorations  extended  over  a  larger  portion  of  Abyssinia, 
I  might  doubtless  have  corrected  my  first  impressions  in  a 
still  larger  number  of  instances.  Von  Heuglin  in  his  later 
papers  generally  gives  the  range  in  elevation  of  each 
species,  and  in  many  cases  his  observations  show  that 
range  to  be  greater  than  appeared  to  me  to^  be  the  case 
in  the  much  more  limited  area  which  fell  under  my 
examination. 

Nevertheless,  I  think  that  the  animals  inhabiting 
Abyssinia  may  very  fairly  be  classed  according  to  the 
elevations   at  which  they  occur.      I   am    disposed  to 


INTRODUCTION,  215 

divide   the   country  hypsometrically   into  four  regions, 
thus: — 

1.  Subalpine  region,  above  9,000  feet.^ 

2.  Temperate  region,  about  6,000  to  9,000  feet. 

3.  Subtropical,  about  3,000  to  6,000  feet. 

4.  Tropical,  below  3,000  feet. 

The  altitudes  are  merely  approximate,  and  vary  some- 
what in  diflferent  localities.  Some  animals  range  through- 
out several  regions  :  thus  the  Hamadryas  monkey,  spotted 
hyaena,  wart-hog,  and  elephant  amongst  mammals,  and 
Neophron  percnopteruSy  the  Lammergeyer,  the  Common 
Kite,  Aquila  rapctx,  Bucorvus  ahyssinicvs,  Pycnonotus 
arsinoey  Corvus  scapulatus,  Buphaga  erythrorhyncha, 
Numida  ptilorhyncha,  and  many  migratory  birds,  are  all 
found  throughout  the  three  lower  regions,  and  some  of 
them  even  in  the  subalpine  zone.  Some  animals  migrate 
with  the  season,  ascending  to  the  highlands  during  the 
summer  and  returning  to  the  lower  elevations  in  winter. 
Others,  such  as  Laniarkts  cethiopicuSf  Coracias  abyssir 
nicay  Corvus  affinis,  LamprocoUus  chalybceitSy  though 
usually  confined  to  the  subtropical  and  temperate  regions, 
are  occasionally  seen  in  considerable  numbers  near  the 
coast. 

The  majority,  both  of  the  mammals  and  of  the  birds, 
range  into  more  than  one  of  the  regions  indicated,  but 
still  in  passing  from  one  to  the  other  a  very  great  and 

1  There  may  possibly  be  a  still  higher  or  Alpine  region  on  the  summits  of 
the  Samyen  mountains,  close  to  the  limits  of  perpetual  snow.  On  a  few  of  the 
very  highest  peaks  the  snow  is  said  never  to  disappear  entirely. 


216  ZOOLOOT. 

marked  change  takes  place,  and  the  great  majority  of  the 
common  birds  are  completely  replaced  by  other  species. 
I  think,  too,  that  the  prominent  birds  and  mammals  ia 
each  region  appertain  to  distinct  sub-faunas  at  least. 
This  will  be  best  shown  by  a  few  brief  remarks  upon  the 
prominent  animals  in  each  region. 

1.  Tlie  Suhalpine  Region. — I  had  only  an  opportunity 
of  examining  this  on  the  Dalanta-Wadela  plateau,  and  in 
two  or  three  of  the  passes  traversed.  My  collections 
were  consequently  very  smalL  The  only  peculiar  mam- 
mals obtained  were  a  Hyrax  and  a  mouse  {Mtis  abys* 
sinicus,  KUpp.),  and  the  following  birds : — 

Pratincda  tordida. 

Saxicola  frenata, 

Macronyx  flavicoUis, 

Pyrrhocorax  alpinus. 

EuplecUs  xantfuymelas.  )  „ 

Crithai^  nigri^p,.       \  ^°"°**  ^^^'^  "  ^«^  *"  »'"^  '***• 

Lohivanellus  melanocephdlns. 

Bemicla  cyanopterti. 

The  following  were  so  abundant  at  10,000  to  10,500 
feet  as  to  be  characteristic  birds,  but  they  were  also  found 
generally  throughout  a  great  portion  of  the  temperate 
region : — 

Buteo  ang^ir. 

Nectarinia  taka^sae., 

Corv^dtur  crmsirostris. 

Chdtuifia  melanoptera. 

Ibis  {Harpijmov)  carunculaiu. 

Anas  flavirostrU. 

The  only  species  in  the  above  list  not  peculiar  to 
Abyssinia  are  the  Alpine  chough  and  Anas  flavirostris : 
the  first  of  which  is  European ;  the  other,  which  is  not 


INTR  OD  UCTION.  2 1  7 

peculiar  to  the  subalpine  fauna,  is  also  found  in  South 
Africa. 

2.  The  Temperate  Region. — This  comprises  the  great 
plateaux  at  7,000  and  8,000  feet,  and  furnishes  the 
typical  fauna  of  Abyssinia.  The  following  mammals 
and  birds  appeared  to  be  amongst  those  most  charac- 
teristic : — 

MAiniALS. 

CaniB  variegatus} 

Antilope  {CephoHophus)  madoqua, 

AntUope  {Scopophorus)  montanxts. 

Lepus  tigrensis. 

Xenis  Imco^twhrinu^, 

Bathyergus  splendens. 

Birds. 


>  Both  fouod  as  low  as  4,500  feet. 


Nisiis  tachiro  (Accipittr  uriduliventery  Riipp.). 

Bubo  cinerascens, 

Turacns  leucotis, 

Colius  kucotis, 

Psiiiacula  iaranta. 

Pogonorhynchiis  uridatus. 

Centropus  manachus. 

Cypsdus  (Bquatorialis. 

Coryihoniis  cristata  (v.  cyanostigina). 

Tockus  Hemprichii,      \ 

Nectarinm  cruentnta.  >  Found  as  low  as  4,500  feet. 

Lanius  humeralis,        ) 

L.  erythropterus.     Found  at  4,000  feet. 

Bradyomis  c}u>€ol<Uina. 

Oriolus  moloxita, 

TurdiLs  sinunsis. 

T,  olivacinus. 

Cossypha  semirufa.     Found  from  4,000  to  9,000  feet. 

Thamnobia  melama, 

Saxicola  lugens. 


*  Most  of  the  manimali*  ranged  to  lower  elevations.     Those  mentioned 
appeared  to  descend  le?s  than  the  others 


218  ZOOLOGY. 

Qraminicola  Levaillantii, 

dtticola  dbyssinica. 

Fanu  dortatus, 

Amydnu  aUnratris, 

A,  BuppeUi,    Found  rarely  at  lower  deyatioDs. 

Hyphantomis  larwUa, 

H.  Churini, 

Colifupasser  laHcauda, 

FringiUa  trutriaUi, 

CrUhagra  strioUUa, 

Columha  dUntarques, 

Turtur  lugm$, 

FterocUi  gtUturalis, 

Franeolinus  Ericelii.    Ranges  from  3,000  to  10,000  feet 

F,  gutturalis.     Eanges  into  the  subtropical  region. 

Otit  nulanogcutir, 

Ballui  Batigdii. 

Of  the  above  list,  whilst  a  considerable  majority  are 
peculiar  to  the  Abyssinian  highlands,  several  are  also 
common  to  Southern  Africa.  Such  are  Nisics  tachiro, 
Bubo  cinerascenSy  Corythamis  cristata,  Graminicola 
Levaillantii,  and  Pterocles  gutturalis. 

3.  The  Subtropical  Fauna. — This  I  met  with  in  the 
pass  leading  to  Senafi^,  and  again  in  the  Anseba  valley. 
The  following  animals  appeared  to  be  the  most  charac- 
teristic : — 

Mammals. 

Cercopithecus  griseo-viridis.    Ranges  into  the  temperate  region  also. 
Canii  mesomdas, 
Sciurui  annulaXug. 

Birds. 
Bvho  lacteiLS, 
Athene  perUUa, 

Chixaerhis  aonura.    Found  also,  but  rarely,  in  the  temperate  region. 
Colius  macrourus. 
Fionus  MeyerL 


12^' TROD  UCTIOX.  219 

PicM  (Bthiopicus. 

Indicator  Sparmanni. 

Pogonorhynchua  abymnicui.    Ranges  into  the  temperate  region. 

Barbaiula  pusUla. 

Oxylophus  afer, 

Cora/sias  pihsa.    Occasionally  seen  in  the  temperate  region. 

Euryatomiis  afer, 

Ispidina  pieta. 

Halcyon  aenegdUnsis, 

Merops  erythropterus, 

M,  Lafresna/yii,    Also  in  temperate  region. 

Tochis  flavirostrii.    In  temperate  region  in  summer. 

T,  eryihrorhynchus.    Also  occasionally  in  tropical  region. 

IrrUor  aUrnmus. 

Nectarinia  affinU.    Ranges  into  the  temperate  region. 

N.pulehella, 

Laniarius  gambenHs, 

DicruTUB  divarieatus. 

Orandala  (or  Pholidauges)  Uueogatter,    Absent  in  winter. 

Oligocercus  micrurus. 

Canuroptera  hrevicaudaia, 

Parus  Imcoptemt, 

Lamprotomia  purpuroptera, 

Hyphantomis  luieola. 

Vidua  principalis, 

Emberiza  Jlaviventrii, 

Treron  ahysnnica,         )  „        ,  .        , 

FrancoHnus  Buppelli.  {  ^^^  "^^  *^*  temperate  region. 

A  great  distinction  will  immediately  be  noticed  in  this 
list.  Instead  of  the  majority  being  peculiar  to  Abys- 
sinia, at  least  two-thirds  range  over  a  large  part  of  Africa, 
being  known  either  from  Senegal  or  the  Cape.  The 
species  also  are,  if  anything,  less  restricted  in  range  than 
those  found  in  the  other  regions. 

4.  The  Tropical  Region. — This  I  only  examined  near 
the  coast.  It  has  a  well-marked  and  peculiar  fauna,, 
consisting  in  a  great  measure  of  desert  species,  but  for 
the  presence  of  which,  indeed,   it  could    ecarcely  be 


220 


ZOOLOGY, 


separated   fi-om   the  last.     The  following   are   charac- 


teristic : — 


¥d\%  rMmxcvXaioL, 
Cants  riparius. 
Chxdla  ScemmertTigi. 
O.  dorau. 


Mammals. 


Btrds. 


Mdierax  polyxonus. 

Pieus  Hempriehii. 

Poganorkynehus  melanocephdltu, 

Cypidus  affinis, 

MeropB  viridiswiMU. 

M.  alhieoUis, 

Todnu  ncLsutus.  Ranging  into 
subtropical  region. 

Nectarinia  habessiniea.  Rang- 
ing into  sabtropical  region. 

N,  metaUica,    Absent  in  winter. 

LaniusfaUax. 

L,  isabdlinuB. 

L,  personatus, 

Laniarius  cruentatus. 

Cereotrichas  erythropterus. 

Saxicola  deserti, 

Cercomela  mdanurct. 


Oryx  beiM, 
Lepus  (eyyptius, 
Xerus  rutUus. 
Dipu9  gerbUlus  Olio, 


DrymcBca  rufifrons, 
D.  gracilis, 
Aidon  galactodes, 
PyrrhtUauda  mdanaucken, 
P,  aJhifrons. 
P,  Uucoiis, 
Chlerita  cristaia. 
Ammomanes  deserti, 
Certhilauda  deseriorum, 
Estrelda  rhodopyga, 
Turtur  aJbiventris, 
Pterodes  exiutus, 
Pt,  Liehtensteini, 
PtemisUs  rubricoUis, 
Otis  arahs, 

(Egialitis  tricollaris, 
Sareiophorus  tectus. 


— ^together  with  the  species  of  gulls,  terns,  and  many 
waders  peculiar  to  the  shores  of  the  Ked  Sea.  Amongst 
the  reptiles  are  species  of  Lacerta  and  Acanthodactyliis 
and  Echis  carinata. 

This  fauna  is  far  less  typically  Abyssinian  than  any  of 
the  others ;  it  contains  but  few  species  which  have  not 
an  extensive  range  in  Northern  Africa,  Arabia,  and 
Persia,  some  being  even  found  in  India. 

Amongst  the  most  striking  instances  of  species  on 
the  highlands  of  Abyssinia  representing  those  in   the 


INTRODUCTWy. 


221 


lower  countiy  arc  the  following.     The  range,  so  far  as 
observed,  is  noted  in  each  instance. 


Lowland  Forms. 


Represbntativu  on  Highlands. 


ft. 
Picu$  Hemprichii     ,  1,000 
Colius  macrourus      .  3,000 
Pogonorhynchus  me- 

lanocephalus     .     .     500 
Centropus  9uperciliO' 

sus 1,000 

Tochis    erythrorhyn- 

chus 1,000 

Nectarinia  hahe^inica  — 
N,  metallica  ...  — 
Saxicola  deserti  .  .  — 
8.  isabellina  ...  — 
Parus  leacopterus  .  4,000 
Crateropus      Uucoce- 

phalus      ....      500 
Hyphantomis      gal- 

bula — 

Turtur  semitorqucUus  1,000 
Lohivanellus  Senegal' 

entis 4,000 


ft 
to  4,000 
„  6,000 

„  3,000 

„  5,rKX) 

„  5,000 
5,000 
2,000 
1,000 
8,000 

„  7,000 


P.  cdhiopicus 
C,  leucotis  .     . 
P.  ahysHnicus 
P.  undoUus  .     . 

C  monachxts   . 

T,  flavirogtris , 
N.  affinis  .  . 
N,  pulcheUa  . 
8,  lugens  .  . 
8.frenata  .  . 
P.  dortaJtui     . 


ft  It. 

3,000  to  6,000 

4,000  „  8,000 

4,000  „  7,000 

.  7,000  „  8,000 

.     6,000  „  8,000 

.     3,000  „  8,000 

.     4.000  „  7,000 

4,000  „  6,000 

.     6,000  „  8,000 

above  10,000 

7,000  to  9,000 


„  4,500  C.  leucopygius.    .     .  4,000  „  8,000 

„  5,000  H.  Gxwrini     .    ,    .  7,000  „  9,000 

„  7,000  T.  lugens    ....  6,000  „  9,000 

„  8,000  X,  mdanocephalus    .  above  10,000 


It  should  be  remembered  that  the  above  range  is 
frequently  exceeded  in  parts  of  the  country  not  visited. 
I  scarcely  think,  however,  that  this  is  anywhere  the  case 
to  a  sufficient  extent  to  vitiate  the  representation. 


222  ZOOLOGY. 


Class  MAMMALIA. 
Order    QUADRUMANA. 
I.  Cynocephalus  hamadryas,  L. 

Hemp,  and  Ehr.  Sjmb.  Phys.  til. 

Riipp.  Keo.  Wirb.  p.  7. — Brehm,  Habesch,  Ko.  2. 

The  great  Dog-faced  Baboon,  the  Sacred  Ape  (Thoth) 
of  the  ancient  Egyptians,  is  by  far  the  commonest 
Monkey  throughout  the  portion  of  Abyssinia  traversed 
by  me.  It  was  met  with  everywhere  from  the  plains 
around  Annesley  Bay  to  the  top  of  the  Dalanta  plateau, 
although  most  abundant,  perhaps,  in  the  tropical  and 
subtropical  portions  of  the  country.  I  saw  a  small  herd 
close  to  Theodore's  old  camp  at  Baba,  on  the  Dalanta 
plateau,  at  above  9,000  feet  of  elevation.  In  the  passes 
leading  to  the  table-land  from  the  coast  immense  num- 
bers were  constantly  seen,  and  the  animals  evidently 
keep  much  to  the  sides  of  rocky  ravines. 

The  herds  vary  in  number ;  some  cannot  include  much 
less  than  250  to  300  Monkeys  of  all  ages.  The  old  males 
are  always  most  conspicuous  animals,  all  the  forepart  of 
their  body  being  covered  with  long  hair.  They  usually 
take  the  lead  when  the  troop  is  moving ;  some  of  them 
also  bringing  up  the  rear ;  others  placing  themselves  on 
high  rocks  or  bushes,  and  keeping  a  sharp  look-out  after 
enemies.     A  troop  collected  on  a  rocky  crag  presents  a 


MAMMAMA.  223 

most  singular  appearance.  I  several  times  saw  large 
numbers  assembled  around  springs  in  the  evening  in  the 
thirsty  Shoho  country  between  Komayli  and  Senate. 
On  such  occasions  every  jutting  rock,  every  little  stone 
more  prominent  than  the  rest,  was  occupied  by  a  patri- 
arch of  the  herd,  who  sat,  with  the  gravity  and  watchful- 
ness befitting  his  grizzled  hair,  waiting  patiently  until 
the  last  of  his  human  rivals  had  slaked  his  thirst  and 
that  of  his  cattle.  Around,  the  females  were  mainly 
occupied  in  taking  care  of  the  young,  the  smaller 
Monkeys  amusing  themselves  by  gambolling  about. 
Occasionally,  if  a  young  Monkey  became  too  noisy  or 
interfered  with  the  repose  of  one  of  his  seniors,  he 
"caught  it''  in  most  umnistakeable  style  and  was  dis- 
missed,  with  many  cuffs,  a  wiser  if  not  a  better  Monkey. 

Cynocephalus  hamadryas  feeds  on  small  fruits, 
berries,  and  seeds,  and  often  (5n  buds  of  trees,  and  on 
young  shoots.  On  the  highlands  I  frequently  saw 
troops  of  them  in  the  fields,  engaged  in  searching  for 
the  "  quentee,"  the  small  tubers  of  Cyperus  esculenttis, 
which  appeared  also  to  be  a  great  resource  of  the  half- 
starved  people  in  Tigrd 

This  species  is  very  rarely  seen  on  trees.  It  appears  to 
avoid  woods,  and  to  keep  mainly  in  the  open  country,  pre- 
ferring, as  already  mentioned,  rocky  precipicea  Hence  its 
habits  differ  entirely  from  those  of  all  the  Indian  Monkeys, 
which  are  tree-loving  animals,  and  indeed  from  Monkeys 
in  general.  It  climbs  heavily  and  clumsily  for  a  Monkey, 
and  when  moving  quickly  on  the  ground  has  a  steady, 
regular  gallop  instead  of  the  bounding  movements  of  a 


224  /AWLOar. 

Presbytes.  Doubtless  the  association  in  such  very  liirge 
herds  is  in  a  great  measure  adopted  as  a  means  of  defence 
against  its  enemies.  From  their  size  and  great  power  of 
jaw  the  old  males  are  most  formidable  antagonists,  and 
their  boldness  in  resenting  injury  is  said  to  be  in  propor- 
tion to  their  power.  Brehm  ('*  Keise  nach  Habesch/'  p.  88) 
relates  an  instance  which  came  under  his  own  notice  of 
their  attacking  a  Leopard  which  had  carried  oflf  one  of  the 
herd,  and  many  stories  are  current  in  Abyssinia  of  their 
attacking  men.  Mr.  Mtinzinger  told  me  that  once  he, 
with  one  or  two  companions,  were  surrounded  by  a  large 
herd,  which  barred  their  path,  and  were  so  threatening 
that  he  was  obliged  to  shoot  one  in  self-defence.  Even 
then,  although  they  fell  back  a  little,  the  Monkeys  did 
not  run  away. 

I  cannot  help  thinking,  however,  that  these  Monkeys 
very  rarely  attack  men,  as  otherwise  some  instances 
would  have  happened  in  the  expeditionary  force,  and  I 
never  heard  of  any.  Near  the  passes  the  flocks  of 
Oynocephalus  soon  became  wary,  as  they  were  frequently 
fired  at  Young  animals,  when  captured,  quickly  became 
tame  and  docile,  but  not  so  much  so  as  the  Cercopithecus, 

d.  Cercopithecus  griseo-viridis,  Desm. 

Desmarest,  Mammalogie,  p.  61. — Brehm.  Habcscfa  No.  1. 

I  met  with  this  monkey  but  rarely.  Unlike  the 
CynocephaluSf  it  is  a  true  tree-monkey,  and  very  rarely 
seen  except  in  forest.  On  the  highlands  I  only  once  saw 
a  flock — this  was  near  Dildi,  south  of  Lake  Ashangi. 
I  met  with  larger  numbers  on  the  AnseLa,  where  they 


MAMMALIA,  225 

inhabited  the  high  trees  on  the  banks  of  the  stream. 
The  flocks  seen  were  small,  not  exceeding  twenty  to 
thirty  individuals.  I  had  but  few  opportunities  of  ob- 
serving their  habits,  but  they  appeared  to  differ  but 
little  from  those  of  Macactis  or  InuuSy  except  that  Cerco- 
piihecn^  is  a  quieter  animal  and  less  mischievous.  In 
captivity  they  are  well  known  as  excessively  docile  and 
good-tempered,  and  fairly  intelligent.^ 

The  following  are  the  dimensions  of  a  full-grown 
female  taken  from  a  freshly-killed  specimen.  The  males 
are  larger.^ 

ft     in. 

Length  firom  upper  lip  to  brow 0    1| 

„     brow  to  nape  over  the  head 0    3| 

„     nape  to  ramp ^    ^\ 

of  the  tail,  20i  ul;  hair  beyond  end,  1^  in.  .    .  I  10 

from  shoulder  to  elbow 0    4^ 

„    elbow  to  wrist 0    4^ 


1  It  is  the  commonest  of  mistakes  amongst  superficial  observers,  and  even 
amongst  naturalists,  to  confound  docility  and  intelligence  in  animals,  and  to 
measure  their  intellectual  powers  by  the  flEicility  with  which  they  can  be 
taught  Hence  the  very  conmion  but,  as  it  appears  to  me,  very  incorrect 
notion,  that  monkeys  are  of  inferior  intelligence  to  such  animals  as  dogs  or 
elephants.  In  reality  they  are  less  docile,  less  willing  to  learn,  and  less 
adapted  to  captivity ;  moreover,  being  of  but  little  use  to  man,  far  less 
trouble  has  been  taken  in  studying  their  habits.  Thus,  while  dog-  and 
elephant-breaking  engage  all  the  time  and  mental  resources  of  particular 
classes  of  men,  the  instruction  of  monkeys  is  left  to  the  unaided  efforts  of 
amateurs  and  organ-grinders.  The  negro  race  amongst  men  appears  to  be  £» 
better  adapted  for  slavery  than  most  savage  races,  being  more  docile  in  a  state 
of  captivity  ;  but  it  is  scarcely  proved  to  be  more  intelligent  on  that  account. 
The  same  reasoning  will  doubtless  apply  to  animals.  I  have  often  seen  dogs 
and  monkeys  kept  together,  and  in  eveiy  instance  it  has  appeared  to  me  that 
the  monkey  ruled  the  dog,  and  that  the  dog  feared  the  monkey,  although  the 
canine  was  the  more  powerful  animal ;  and  I  can  only  account  for  this  by  the 
superior  intelligence  of  the  monkey. 

All  measurements  are  in  English  feet  and  inches. 

Q 


226  EOOLOGY. 

ft  in. 

Length  of  hand 0  3 

„    from  hip  to  knee 0  5 

„      „     knee  to  heel 0  6 

„    of  hind  foot 0  4j 

Height  of  ear 0  If 

Breadth  of  ear 0  If 

Cynocephalus  hamadryas  and  Cercopithecus  griseo- 
viridis  were  the  only  two  monkeys  of  which  specimens 
were  obtained.  I  saw  a  third  species,  perhaps  Thero- 
piihecus  ohscurus,  Heugl.,  at  Magdala.  The  beautiful 
Colohus  griereza,  of  which  skins  are  frequently  brought 
to  Aden  from  the  mountains  of  the  Somali  country,  I 
never  heard  of  while  in  Abyssinia. 

Ordrb  cheiroptera. 

3.  Vespertilio,  sp. 

Bats  were  remarkably  scarce  throughout  the  Abys- 
sinian highlands,  partly  perhaps  owing  to  the  season  of 
the  year,  and  I  only  obtained  one  specimen,  which  is  in 
too  poor  condition  for  determination. 

Order  CARNIVORA. 

4.  Fells  maniculata,  Kiipp. 

Riippell,  Adas,  t.  19. 

jP.  caligata,  partim,  Gray,  Cat.  Cam.  &c.  Mamm.  in  Brit.  Mua. 
1869,  p.  29. 

I  obtained  a  single  specimen  of  this  cat  at  Zoulla.  Its 
colour  is  pale  rufous  grey,  more  rufous  and  rather  darker 
in  the  centre  of  the  back,  with  indistinct  spots  on  the 
flanks.     Ears   rather  deeper  rufous,   the  hairs   at  the 


MAMMALIA.  227 

extremity  very  slightly  lengthened.  Legs  with  more 
distinct  but  rather  irregular  transverse  stripes,  which 
become  darker  below :  one  just  below  the  elbow  on  the 
fore  leg  is  black,  and  particularly  distinct  on  the  inside. 
Soles  of  feet  black.  Tail  ringed  towards  the  tip,  the 
last  ring,  and  the  tip  for  about  1^  inch,  black:  alto- 
gether three  rings  are  distinguishable  besides  the  tip, 
each  one  less  distinct  than  that  behind  it  and  at  a 
greater  distance  from  the  last.  Hairs  of  the  back  when 
examined  singly  are  rufescent  grey  with  a  black  tip, 
the  latter  frequently  wanting  on  the  flanks.  Head 
about  3  inches  long,  from  between  ears  to  insertion 
of  tail  about  18  inches,  tail  12  inches.  The  measure- 
ments, being  taken  on  the  dried  skin,  are  not  quite  accu- 
rate, with  the  exception  of  the  tail,  although  very  nearly 
so.  None  of  the  hairs  on  the  back  much  exceed  1^ 
inches  in  length. 

Although  the  "  Monographies  "  of  Temminck  are  dated 
1827  on  the  title-page,  and  the  Atlas  to  Ruppeirs 
"Reise"  1826,  the  former  imquestionably  appeared  the 
first,  as  in  the  latter,  in  the  description  of  F.  mani- 
culata,  reference  is  made  to  the  page  in  Temminck 
in  which  it  is  mentioned  as  a  new  species  found  by 
Ruppell.  If  therefore,  as  Dr.  Gray  considers,  F.  mani- 
culata  is  identical  with  Temminck's  F.  caligata,  he  is 
quite  correct  in  upholding  the  latter  name.  It  should, 
however,  be  borne  in  mind  that  the  two  are  men- 
tioned by  Temminck  as  distinct,  and  I  am  disposed 
to  consider  Temminck's  animal  to  be  the  following 
species. 

q2 


228  ZOOLOGY, 

5.  F.  caligata,  Temm. 

Booted  Lynx  of  Brace,  Travela,  vol.  v.  p.  146,  with  plate. 
¥,  caligata,  Temm.  Monographies  de  Mammalogie,  L  p.  123. 

It  is  extremely  difficult  to  ascertain  the  correct  specific 
names  for  cats.  Dr.  Gray  may  perhaps  be  right  in 
uniting  all  these  African  forms.  In  this  case  as  in 
hundreds  of  others,  if  a  large  series  be  examined,  it  wiU 
be  found  that  there  is  a  gradation  between  the  extreme 
forms,  and  yet  the  Grordian  knot  is  only  cut  by  classing 
all  together.  In  the  present  instance,  the  single  speci- 
men of  a  wUd  cat  which  I  obtained  from  the  highlauds 
of  Abyssinia  is  so  entirely  different  fix)m  the  specimen 
lafit  described,  that  no  naturalist  looking  at  these  two 
only  would  consider  them  to  belong  to  the  same  species. 
The  specimen  which  I  assign  to  F.  caligata  was  shot  at 
Adigrat  at  an  elevation  of  about  8,000  feet  above  the 
sea.  It  is  at  least  twice  the  size  of  the  specimen  of 
F.  maniculata ;  the  measurements,  again  unfortunately 
only  taken  from  the  dried  skin  (which,  however,  is  in 
beautiful  order  and  in  no  way  distorted),  being — 

ft    in. 

Head  and  body  to  insertion  of  tail 2    1 

TaU 11 

Total 3    2 

The  head  is  about  4  inches  long,  ears  2  in.  long,  radius 
5^  in.,  carpal  joint  to  end  of  toe  3  in.,  tibia  5^  in.,  tarsus 
and  hind  toes  5^  in.  The  general  colour  is  a  somewhat 
rufous  grey,  darker  from  being  mixed  with  black  towards 
the  centre  of  the  back.  The  hair  of  the  fur  is  fulvous, 
becoming  paler  on  the  flanks,  dusky  at  the  root,  grey 


MJMMJLIjL  229 

towards  the  tip,  with  a  short  dusky  ring  between  the 
isabelline  and  grey,  and  more  or  less  of  a  black  tip,  espe- 
cially towards  the  centre  of  the  back.  The  hairs  on  the 
back  are  about  2^  inches  long.  Under  parts  rufous 
white.  The  tip  of  the  tail  for  about  2  inches  is  black  : 
there  is  a  black  ring  about  an  inch  further  forward,  and 
indications  of  a  second.  The  cheek  and  face-markings 
are  very  faint.  Ears  rufous  brown,  scarcely  pencilled  at 
the  end.  There  is  a  faint  black  ring  round  the  shoulder, 
much  more  strongly  marked  inside,  as  usual.  The  soles  of 
all  the  feet  are  black. 

This  cat  agrees  exactly  both  in  the  character  of  the  fiir 
and  in  coloration  with  Cape  specimens,  and  there  are 
skins  both  in  the  British  Museum  and  in  that  of  Berlin 
which  are  precisely  similar.  This  is  an  important  point, 
because  in  this  instance,  as  in  others,  the  animal  from  the 
lowlands  of  Abyssinia  is  a  tropical  form  and  similar  to 
others  found  in  the  neighbouring  tropical  countries, 
whilst  that  from  the  highlands  is  a  form  found  again 
in  a  distant  part  of  Africa,  where  the  climate  is  cooler. 
This  fact  is  of  more  interest  than  the  question  whether 
these  various  races  shall  be  called  species  or  varieties. 

I  have  scarcely  a  doubt  that  this  is  the  *' Booted 
Lynx"  of  Bruce,  and  Temminck  especially  identifies 
Brace's  description  with  his  F.  caligata.  In  Brace's 
figure  the  ears  are  much  too  large,  and  too  strongly  peni- 
cillated,  as  remarked  by  Temminck.  Indeed  this  cat 
scarcely  belongs  to  the  lyncine  group. 

At  Berlin  I  found  this  cat  labelled  F.  caffra,  Desm.  : 
but  it  does  not  coincide  with  Desmarest's  description. 


230  ZOOLOGY. 

6.  Felis  leo,  L. 

Although  no  specimens  were  brought  away,  a  few 
words  on  the  Abyssinian  lion  may  not  be  out  of  place. 
The  sportsmen  of  the  army  who  went  to  Abyssinia  with 
the  expectation  of  shooting  large  game  were  greatly  dis- 
appointed ;  on  the  route  selected  the  paucity  of  large 
animals  was  remarkable.  Not  a  single  lion  was  shot  by 
any  officer  of  the  expedition,  and  only  one  or  two  were 
seen.  In  my  subsequent  journey,  however,  to  the  Anseba, 
I  met  with  several,  as  related  above,  and  aided  in  killing 
one  lioness.  The  incident  has  been  related  in  a  previous 
page,  and  is  worthy  of  notice,  as  it  illustrates  the  bold- 
ness of  the  Uons. 

On  the  Anseba  lions  aboimded,  and  the  valley  re- 
soimded  with  their  roaring  every  night.  They  were  not 
often  seen,  for  they  usually  appeared  to  pass  the  day  in 
the  dense  thickets  on  the  river-banks,  or  in  the  neigh- 
bouring ravines,  and  they  were  often  heard  roarmg  in 
these  thickets  towards  evening.  In  one  instance  Captain 
Mockler  and  I  listened  to  one  for  some  time,  nearly  an 
hour  before  sunset.  It  was  useless  for  us  to  attempt  to 
shoot  him  in  the  thicket  in  which  he  then  was.  The 
Bedjuk  people  who  were  with  us  assured  us  they  could 
bring  him  out  into  the  open.  These  people  firmly  believe 
that  the  lion  understands  them  when  they  speak  to  him, 
and  that  if  they  abuse  him  sufficiently  to  make  him 
angry  he  will  come  out  to  attack  them.  The  plan  for 
this  hovel  lion-hunt  therefore  was  that  all  should  stand 
outside  the  thicket,  and  that  the  Bedjuk  should  call  him 
names.     I  regret  to  say  that  the  experiment  was  unsuc- 


MAMMALIA.  231 

cessfiil,  the  whole  vocabulary  of  Tigr^an  Billingsgate  failed 
in  arousing  the  wrath  of  his  majesty  sufficiently  to 
induce  him  to  make  his  appearance ;  and  as  night  came 
on,  and  we  retreated  homewards,  he  roared  more  loudly 
than  ever  in  the  depths  of  his  thicket.  Like  all  true 
believers,  our  Bedjuk  friends  were  by  no  means  shaken 
in  their  faith  or  inclined  to  doubt  that  the  lion  had 
understood  their  abuse ;  in  fact,  their  opinion  of  his  intel- 
ligence  was  increased  by  the  circumstance  that  he  would 
not  venture  out  while  there  were  guns  in  the  way ;  and 
they  assured  us  that  had  they  been  alone  he  would  have 
attacked  them  for  daring  to  abuse  him. 

Lions  kill  many  cattle,  mules,  donkeys,  and  horses  in 
the  Habale  and  Samhar,  and  are  said  to  migrate  with  the 
herds,  following  them  to  the  lowlands  in  the  winter,  and 
ascending  with  them  to  the  mountains  in  summer.  Men 
are  often  killed  by  them  :  we  heard  of  several  instances 
taking  place  during  our  short  stay. 

Having  had  many  opportunities  of  observing  the 
Indian  tiger,  I  may  make  a  few  remarks  on  the  diflfer- 
ences  between  the  habits  of  the  two  great  carnivores. 

The  first  peculiarity  that  struck  me  in  the  African  lion 
was  their  noisiness.  I  have  constantly  been  for  months 
together  in  countries  in  India  abounding  in  tigers  with- 
out hearing  their  cry.  Indeed,  it  is  by  no  means  a 
common  sound  in  an  Indian  forest;  leopards,  I  should 
say,  are  much  more  frequently  heard  than  tigers.  The 
cry  of  the  two  animals,  commonly  known  as  roaring, 
though  it  is  utterly  different  from  the  harsh  growl  of 
anger  to  which  the  term  might  most  appropriately  be 


232  ZOOLOGY. 

applied,  is  very  similar,  and  consists  of  several  deep 
notes  uttered  rather  quickly  one  after  the  other,  and 
repeated  at  longer  or  shorter  intervals.  I  never  heard 
tiffers  until  it  was  nearly  dark  As  already  mentioned, 
ul  ««  Bometime.  .^  noby  before  d«=k 

Another  characteristic  of  lions  is  their  greater  bold- 
ness. The  attack  upon  our  camels  in  open  day,  near 
Ain,  and  the  circumstance  of  a  lioness  lying  in  bushes 
within  200  yards  of  our  camp,  in  spite  of  shots  fired  at 
her,  at  Ailat,  are  instances.  I  have  known  of  similar 
cases  with  tigers,  but  they  are  very  unusual. 

Both  animals  resemble  each  other  in  their  wandering 
habits,  and  in  their  preference  for  sandy  roads  or  river- 
beds for  their  nocturnal  rambles,  the  feet  in  both  being 
too  tender  to  stand  much  walking  over  sharp  stones. 
The  largest  tracks  of  lions  seen  by  us  were,  as  nearly  as 
I  could  tell,  about  equal  in  size  to  those  of  the  largest 
tigers. 

I  only  saw  one  cow  which  had  been  killed  by  a  lion, 
and  in  this  instance  the  neck  was  not  broken,  as  it  almost 
invariably  is  in  large  animals  killed  by  tigers. 

The  Abyssinian  lion  has  only  a  short  mane  like  that 
of  Senegal,  and  is  consequently  considered  by  some 
naturalists  as  a  distinct  race  from  the  heavily-maned 
Cape  and  Barbary  lions.  The  Persian  lion,  figured  in 
the  Revue  et  Magasin  de  Zoologie  for  1862,  shows  the 
same  peculiarity.  The  Indian  lion  has  now  been  clearly 
proved  to  have  a  long  mane  when  adult ;  the  maneless 
animals,  on  which  the  variety  Felis  leo  goojeratensis 
was  founded,  having  been  shown  to  be  young  animals. 


MAMMALIA.  233 

This  irregular  geographical  distribution  of  the  two 
varieties  is  opposed  to  the  idea  of  their  being  really 
distinct  forms. 

The  following  dimensions  of  the  lioness  killed  by 
Captain  Mockler,  near  Ain,  were  taken  on  the  carcase 
about  an  hour  after  death,  and  before  skinning.  The 
lions  seen  by  us  were  larger  animals,  but  this  appeared 
to  be  a  full-sized  female. 

ft    in. 

Length  of  face,  firom  the  nose  to  between  the  ears  ..11 
„  from  between  ears  to  top  of  shoulder  ....  1  3 
„       of  back  from  the  top  of  the  shoulder  to  the 

insertion  of  the  tail 3     1 

„       of  tail,  including  the  hairs  at  the  end  ....    2    7 

Total  length  from  nose  to  end  of  tail,  measured  along  the 

curve  of  the  back  ^ 80 

Length  of  the  body  measured  in  a  direct  line  along  the 

side  from  the  front  of  the  shoulder  to  the  rump      .  3    6 

Height  at  shoulder 2     8 

Depth  of  body  behind  shoulders 13^ 

Length  of  the  ear 04 

Girth  of  chest  behind  shoulders 3    3 

neck 19 


yy 


» 


head  in  front  of  the  ears 2     1 


„       forearm 11 


7.  Genetta  tigrina,  Schrober. 

Fivcrra  tigrina,  Schreb.  Saugth.  III.  p.  425,  t.  1 15. 

O.  abymnica,  Riipp.  Neu.  Wirb.  p.  33,  t.  11. 

G,  tigrina.  Gray,  Cat.  Cam.  &c.  Mamm.  Brit.  Mus.,  1869,  p.  51. 

I  am  indebted  to  Lieut.  St.  John,  who  had  charge 
of  the   telegraph   in  Abyssinia,  for  a  skin  of  a  civet- 

^  This  is  the  usual  sportsman's  plan  of  measuring  wild  animals,  and  is  less 
liable  to  inaccuracy  than  measuring  in  a  straight  line.  The  skin,  when  taken 
off  and  stretched,  always  measures  one-eighth  to  one-quivrter  more. 


234  ZOOLOGY, 

cat,  shot  near  Annesley  Bay  I  believe,  which  I  can 
only  refer  to  this  species.  If  Dr.  Gray  is  correct  in 
uniting  the  various  races  from  different  parts  of  Africa 
under  G.  tigrinay  I  think  there  can  be  but  little  doubt 
but  that  the  present  specimen  must  also  belong  to  the 
same.  It  diflfers  materially  from  Rlippell's  figure  of 
G.  ahyssinica,  there  being,  including  the  tip,  only  ten 
black  rings  on  the  tail  instead  of  eighteen. 

There  is  an  Abyssinian  specimen  in  the  British  Museum 
labelled  G.  amer.  Rtipp.,^  which  is  considered  by  Dr.  Gray 
to  be  a  variety  of  G.  tigrina.  It  differs  from  my  speci- 
men in  being  of  a  darker  colour,  and  in  the  bands  on 
the  flanks  not  being  broken  into  spots.  A  young 
specimen  in  the  British  Museum  labelled  G.  abyssinica, 
from  Sennaar,  has  nine  taU-rings.  The  markings  are 
very  similar  to  those  in  my  specimen, 

8.  Herpestes  mutgigella,  EUpp. 

Riipp.  Neu.  Wirb.  p.  29,  t.  9,  f.  1. 

To  this  species  I  refer  all  the  specimens,  seven  in 
number,  collected  by  me  in  Abyssinia.  All  were  ob- 
tained from  Senafd  or  Adigrat^  except  one  specimen 
from  Sum.  They  vary  much  in  colour,  some  being 
very  much  more  rufous  than  others.  Usually  the  hinder 
part  of  the  back  is  more  or  less  rufescent.  Some  are 
deep  rufous  brown  along  the  whole  back  and  on  the 
back  of  the  head ;  all  have  the  tail  tipped  with  black. 
The  greyest  specimen  of  all  is  that  from  the  Suru  pass, 

^  Evidently  a  MS.  name  merely. 


MAMMAUA.  235 

shot  at  an  elevation  of  only  2,000  feet  above  the  sea, 
and  I  at  first  thought  this  might  be  JT.  gracilis^  Rtlpp. ; 
but  it  has  the  black  tail-tip  of  H.  mutgigella.  I  have 
seen  a  precisely  similar  specimen  in  the  Berlin  Museum, 
labelled  H.  gracilis^  and  I  much  doubt  if  the  two  forms 
are  really  distinguishable.  The  tail  varies  in  length  in 
diflferent  specimens  from  twelve  to  fifteen  inchea 

This  ichneumon  was  usually  seen  in  low  bush  jungle ; 
often  upon  rocks.  It  is  diurnal  in  its  habits.  I  also 
once  or  twice  saw  a  striped  species,  probably  H. 
zehniy  RUpp.,  of  which  I  did  not  succeed  in  securing 
specimens. 

g.  Hyaena  crocuta,  L. 

If.  crocuta  ct  If.  maculata,  auct. 
Crocuta  maculata,  Gray. 

From  the  time  of  entering  Abyssinia  to  that  of 
leaving  it,  the  singularly  unearthly  howl  of  the  spotted 
hyaena  was  the  commonest  of  nocturnal  sounds.  Yet  it 
was  the  rarest  occurrence  to  see  one  of  these  a.nima1s 
in  the  day.  I  only  turned  up  two  or  three  fi-om  ravines 
and  thickets,  and  occasionally  caught  sight  of  one 
prowling  about  towards  dusk.  At  night  the  hyaenas 
constantly  came  amongst  the  tents,  and  sometimes 
attacked  goats,  mules,  ponies,  or  cattle,  which  were 
tethered  near  the  camp.  They  were  just  as  abundant 
on  the  highlands  as  in  the  low  coimtry. 

In  boldness  and  rapacity  the  spotted  hyaena  is  far 
superior  to  his  striped  relative,  so  far  at  least  as  can  be 
judged  from  the  habits  of  the  latter  in  India.     I  have 


236  ZOOLOGY. 

never  known  HycBna  striata  to  kill  any  animal  larger 
than  a  goat,  and  it  is  mainly  a  carrion-feeder.  In  the 
Abyssinian  campaign  I  met  with  several  instances  of 
both  cattle  and  mules  being  attacked  by  spotted  hyaenas. 
A  full-grown  cow  just  outside  our  camp  at  Ailat  was 
torn  to  pieces  and  entirely  devoured  by  hysenas  before 
morning.  These  animals  evidently  collect  in  consider- 
able numbers.  In  India  it  is  a  common  practice  amongst 
sportsmen  to  tie  up  young  buffaloes  or  bullocks  as  baits 
for  tigers,  in  places  haunted  by  those  animals.  In  hun- 
dreds of  instances  in  which  I  have  known  this  done, 
I  never  remember  one  of  the  baits  being  killed  by 
hysBuas,  although,  after  a  bullock  or  buffalo  is  killed 
by  a  tiger  or  leopard,  hyaenas  almost  always  feed  on 
the  carcase. 

The  striped  hyaena  is  a  noisy  animal  at  night,  but  far 
less  so  than  H.  crocutay  and  there  is  a  considerable 
difference  in  the  voice. 

Striped  hyaenas  are  said  to  be  sometimes  seen  in  parts 
of  Abyssinia,  and  I  heard  from  Mr.  Miinzinger  that  a 
third  smaller  animal  occurs  near  Massowa,  which  is  much 
dreaded  by  the  natives,  as  considerable  numbers  of  them 
collect  and  attack  men.  This  may  be  Lycaon  picta,  and 
not  a  hyaena. 

In  gait  Hycena  crocuta  is  much  less  ungainly  than 
H.  stiiata,  the  disproportion  ia  the  length  of  the  legs 
being  much  smaller.^ 

^  In  the  Catalop;ue  of  Carniyorous,  &c.  Mammalia  in  the  British  Mugeum, 
1869,  Hyaena  as  restricted  is  said  to  have  the  legs  subequal,  and  CroctUa  to 
have  the  hinder  legs  short    These  characters  8h(Hild  be  reversed. 


MAMMAUA.  237 

The  following  axe  the  dimensions  of  a  fine  male  killed 
at  Takonda,  near  the  head  of  the  Haddas  pass  : — 

ft  in. 

Length  of  head  from  nose  to  nape 0  11 

„        neck  and  body  firom  nape  to  romp,  measured 

along  the  cnnre  of  the  back 3      1 

„        tail,  I  ft. ;  hairs  beyond  end,  6  in.    ....     1      6 

Total  length  firom  nose  to  end  of  tail,  measured  along  the 

coire  of  the  back 6      6 

Height  at  shoulder 2      6 

Length  of  body  from  shoulder  to  rump,  measured  along 

the  side 2      6 

Girth  of  neck 1  10 

„        chest  behind  shoulders 2  11 

„        forearm 0      9^ 


10.  Canis  mesomelas,  Gm. 

Syst.  Nat.  I.  pt.  1,  p.  73. — Riipp.  Neu.  Wirb.  p.  39. 

This  very  handsome  jackal  was  occasionally  seen  in 
the  pass  between  Komayli  and  Senaf^,  and  much  more 
frequently  about  the  base  of  the  hills  west  of  Massowa, 
and  on  the  Anseba.  I  believe  that  I  met  with  it  also 
in  tlie  desert  country  north-west  of  Massowa.  On  the 
shores  of  Annesley  Bay  it  was  never  observed. 

The  following  dimensions  are  those  of  a  male  animal 
killed  in  the  Anseba  valley  : — '- 

ft     in. 

Length  from  nose  to  between  ears 0    6} 

„        „      between  ears  to  top  of  shoulder  ....  0    5^ 

„        „      top  of  shoulder  to  rump 12 

„      of  tail,  1  ft.  2  in. ;  hairs  beyond  end,  2  in.     .     .  1     4 

Total  length  from  nose  to  tip  of  tail  measured  along 

the  curve  of  the  back 3    6 


238  ZOOLOGY. 

ft.     n. 

Length  of  body  from  ahoulder  to  rump,  measured  along 

the  side 18 

Height  at  shoulder 16 

Girth  of  neck 0  10 

,y     body  behind  shoulder 1     3| 

On  one  occasion,  at  least,  animals  of  this  species 
were  seen  in  the  immediate  vicinity  of  a  lion.  In 
fact,  it  was  their  peculiar  movements,  walking  slowly 
away  with  constant  glances  towards  a  particular 
bush,  that  drew  our  attention  to  the  place  where  the 
lion  lay. 

ff 
II.  C.  variegatus,  Biipp. 

Riipp.  Atlas,  1. 10.— Neu.  Wirb.  p.  39. 

This  is  the  common  jackal  of  the  Abyssinian  high- 
lands. It  may  also  occur  at  low  elevations,  but  of  this 
I  am  not  certain.  By  RUppell  and  Brehm  it  is  con- 
sidered merely  a  variety  of  the  last  species  ;  but  in  this 
view  I  cannot  concur,  as  amongst  the  numerous  jackals 
seen  by  me  on  the  highlands  I  never  recognised  a 
specimen  of  C.  mesomeJas,  which  I  never  met  with 
above  about  5,000  feet  I  also  think  the  present  a 
larger  animal.  The  measurements  given  below  only 
convey  a  very  faint  idea  of  the  proportionally  greater 
weight  and  bulk  ;  in  the  dried  skins  the  limbs  of 
C.  variegatiLS  are  very  much  larger  than  those  of 
C.  mesomelas. 

I  must  express  my  surprise  at  finding  these  African 
jackals — for  they  are  true  jackals — classed  in  Dr.  Gray's 
list  with  foxes,  whilst  the  Indian  jackal,  Ca^iis  aurev^y 


MAMMAUJ.  239 

is  placed  with  the  wolves  (P.  Z.  S.  1868,  pp.  504, 516,  and 
Cat.  Carn.  &c.  Mamm.  Brit.  Mus.  1869,  pp.  188,  203). 
If  the  genus  Canis  is  to  be  subdivided  to  the  extent 
which  is  done  by  Dr.  Gray,  the  jackals  {Saccalius)  are 
a  much  more  natural  group  than  some  of  those  adopted, 
and  in  all  their  habits  diflfer  greatly  from  foxea  They 
are  far  more  social,  and  are  eminently  scavengers, 
feeding  upon  carrion,  and  haunting  the  neighbourhoods 
of  towns  and  villages.  Their  cry  is  peculiar  and  pro- 
longed, never  a  sharp  bark  like  that  of  a  fox.  Their 
build  is  quite  different ;  the  tail  and  ears  are  shorter,  the 
legs  longer,  the  muzzle  blunter,  and  the  jaws  and  teeth 
much  more  powerful.  They  never  use  burrows  in  the 
earth,  except  when  they  have  yoimg.  I  unfortunately 
omitted  to  notice  the  form  of  the  pupil  in  C.  mesomelas 
and  C.  variegatus:  it  is  doubtless  round,  as  in  C 
aureus. 

Some  specimens  of  C.  variegatus  resemble  some  of 
C.  aureus  so  closely,  that  I  doubt  if  the  two  forms 
could  be  distinguished  by  their  skins.  The  former 
is  a  larger  animal,  however,  and  there  is  an  important 
difference  in  the  cry.  That  of  C.  aureus  begins  with 
a  long  shrill  howl,  repeated,  with  a  slight  rise  in 
the  scale  at  the  commencement,  three  or  four  times, 
and  followed  by  a  modulated  series  of  short  barks. 
It  is  best  conveyed  by  the  well-known  version,  "  Dead 
Hindoo-oo-oo,  where,  where,  where  ? "  both  the  first 
and  second  part  being  repeated  separately.  Now,  in 
the  cry  of  the  Abyssinian  jackal,  the  second  portion, 
expressed  by  "  where,  where,  where,"  is  entirely  omitted. 


240  ZOOLOGY. 

and  only  the  long  wailing  howl,  a  shrill  representation 
of  the  noise  made  by  a  dog  when  "baying  the  moon/' 
is  uttered. 

The  following  are  the  dimensions  of  two  very  fine 
Abyssinian  jackals,  male  and  female,  shot,  the  first  close 
to  Senafd,  the  second  at  Halai : — 

Male.       Female, 
ft    in.      ft    in. 

Length  from  nose  to  between  ears 0    7|    0    7j^ 

„       „     between  ears  to  rump 2    0      1  11 

„      of  tail,  10  in. ;  hairs  beyond  end,  3  in.     .     .     1     1       10} 

Total  length  from  nose  to  tip  of  tail,    measured 

along  the  curve  of  the  back 3    8}    3    7 

Length  of  body  from  shoulder  to  rump,  measured 

along  the  side 18        — 

Height  at  shoulder 18^16 

Girth  of  neck —  0  10 

„      body  behind  shoulder —  1    ^i 

Length  of  ear —  0    6 

Longest  hairs  in  moustache 0    3^      — 

The  skull  of  the  male  above  mentioned  is  6 '7  inches 
long,  3*4  wide  at  the  zygomata,  and  2'75  high  measured 
through  the  front  of  the  orbit^ 

About  Annesley  Bay  there  was  a  small  kind  of 
jackal,  with  long  legs  and  longish  ears,  of  a  rather 
pale  sandy  colour,  and  very  slight  build.  I  never  suc- 
ceeded in  obtaining  a  specimen.  It  was  probably  Cards 
riparius^  Hempr.  and  Ehr.,  described  in  the  "  SymbolaB 
Physicae." 

^  The  skull  of  an  old  specimen  of  C  aurmSy  in  the  British  Museum, 
measures  in  inches, — ^length,  5*75 ;  breadth,  3*1  ;  height,  2*7 ;  and  the 
teeth  are  much  smaller  than  in  C,  variegcUuSj  which  is  evidently  a  larger  and 
more  powerful  animal. 


MAMMALIA.  241 


Order  PACHYDERMATA. 
za.  Phacochoerus  ^liani,  Kiipp. 

Phacochoerus  JEliani,  Riipp.  Atlas,  t.  25,  26. 
P.  harojcL,  Hemp,  and  Ehr.,  Symb.  Phys.,  t  20. 
P.  cBthxopieuSf  par  tint,  Gray,  P.  Z.  S.  1868,  p.  46. 

The  wart-hog  occurs  throughout  Abyssinia,  if,  as  I 
have  every  reason  to  believe,  the  species  occurring  on  the 
highlands  is  the  same  as  that  found  at  lower  elevations. 
Near  Annesley  Bay  it  abounds,  and  I  shot  several  fine 
specimens. 

Its  habits  are  very  similar  to  those  of  ordinary  pigs. 
It  lives  amongst  bushes  or  in  ravines  during  the  day,^ 
and  comes  out  to  feed  in  the  evening,  still  keeping  much 
to  bush  jungle.  The  large  males  are  usually  solitary  ; 
the  younger  animals  and  females  live  in  small  herds, 
apparently  not  exceeding  eight  or  ten  in  number.  I 
never  saw  large  "  sounders,"  such  as  are  so  commonly 
met  with  in  the  case  of  the  Indian  hog.  It  feeds  much 
on  roots,  which  it  digs  up  by  means  of  its  huge  tusks. 
It  also  appears  to  dig  large  holes,  in  which  it  occasionally 
lies.     These  are  perhaps  intended  for  the  young. 

Despite  its  formidable  appearance,  the  Abyssinian 
wart-hog  is  a  comparatively  timid  animal,  far  inferior 
in  courage  to  the  Indian  wild  hog.  Several  which  I 
wounded  showed  no  inclination  to  charge  under  circum- 

1  In  p.  Z.  S.  1868,  p.  45,  Phdcochcerus  is  said  by  Dr.  Gray  to  live  floating 
amongst  reeds.  This  is  certainly  not  the  case  with  the  Abyssinian  species, 
nor,  so  fjBir  as  I  am  aware,  with  the  South  African  animal  generally. 

R 


242  zooLoor. 

stances  in  which  an  Indian  pig  would   certainly  have 
shown  fight. 

The  flesh  is  savomy,  but  dry  and  hard,  even  in  com- 
paratively young  animals. 

Elian's  wart-hog  is  considered  by  Dr.  Gray,  the  last 
writer  on  the  subject,  identical  with  the  South  African 
Phacochcerus  cethiopicus,  Erxl.  {Sus  africanuSy  Gm.)  ; 
and  unquestionably  Dr.  Gray  gives  very  strong  arguments 
in  favour  of  his  views.  My  reasons  for  dissenting  are  : 
first,  that  Dr.  Gray,  from  the  nature  of  the  specimens 
before  him,  was  compelled  to  rely  entirely  upon  cranial 
and  osteological  characters,  and  that  I  believe  there  are 
external  differences  in  the  two  forms  ;  secondly,  that 
all  the  specimens  in  the  British  Museum  appear  to  be 
from  South  or  West  Africa  and  not  from  Abyssinia  ; 
and  it  is  most  probable  that  the  two  races  are  repre- 
sentative. 

The  external  differences  are  far  from  unimportant,  as 
I  judge  from  a  fine  adult  South  African  specimen  in  the 
Gardens  of  the  Zoological  Society,  and  from  the  drawing 
in  Major  Comwallis  Harris's  "  Game  Animals  of  South 
Africa."  The  ears  in  the  South  African  species  have 
tufts  of  long  hair ;  this  peculiarity  is  wanting  in  P. 
^liani:  and  the  form  of  the  warty  protuberances 
on  the  face  is  quite  different  in  the  two  species, — in 
the  males,  at  all  events  ;  in  the  females  the  warts  are 
much  smaller,  and  the  upper  ones  ai-e  deficient.  The 
colour  of  P.  CBthiopicus  also  appears  much  more  rufous 
in  general. 

So  far  as  regards  the  dental  chai-actcrs,  all  the  evi- 


MAMMALIA.  243 

dence  which  I  can  add  is  in  favour  of  the  distinction. 
I  brought  away  with  me  three  skulls  of  large  males. 
One  is  that  of  a  very  old  animal  with  enormous  tusks, 
the  upper  pair  being  no  less  than  eleven  inches  long 
each  outside  the  jaw,  measured  along  and  outside  of 
the  curve.  In  all  three  skulls  both  upper  and  lower 
incisors  are  permanent.  In  old  specimens  of  P.  cBthio- 
picus  both  are  generally  wanting.^ 

13.  Rhinoceros  keitloa,  A.  Smith  (R.  bicamis,  L.  var.). 

Bruce,  Travels,  voL  v.  p.  85. 

Smith,  Illust.  S.  Afr.,  Zool.-Mamm.  pi.  1. 

Bhinast^  keitloa.  Gray,  P.  2.  S.  1867,  p.  1025. 

Khinoceroses  appear  to  be  confined  to  the  lower 
elevations  in  Abyssinia ;  they  do  not  ascend  above 
about  5,000  feet  elevation  in  the  northern  portion  of 
the  country,  and  they  are  entirely  wanting  on  the  high 
plateau.  I  only  met  with  them  on  the  banks  of  the 
Anseba,  where  they  were  tolerably  common.  AU  be- 
longed to  the  black  type — the  white  African  rhinoceros 
does  not  appear  to  have  been  met  with  north  of  the 
equator. 

Although  the  rhinoceros  of  Abyssinia  and  Nubia  was 
described  long  ago  by  Bruce,^  very  few  specimens  have 
ever  reached  Europe.  The  diflSculty  of  carrying  the 
spoils  of  so  large  an  animal  is  very  great  Mr.  Jesse 
brought  away  the  complete  skeleton  of  an  adult  female, 

^   Vide  also  Sclater,  Proc.  Zool  Soc.  1869,  p.  276,  pL  xx. 
'  Bruce's  figure,  however,  represents  an  Indian  (one-horned)  Rhinoceros, 
with  an  additional  horn. 

R    2 


which  is  now  in  the  British  Museum.  I  only  obtained 
the  head  and  skull  of  a  younger  female,  nearly  adult, 
now  deposited  in  the  same  collection.  The  animal  is 
by  no  means  easy  to  kill,  and  these  two  were  the  only 
ones  bagged  by  our  party.  There  ia  now  a  young  male 
from  Nubia  in  the  Gardens  of  the  Zoological  Society.^ 


The  accompanying  figures  represent  the  head  and 
horns  of  the  yoimger  female  killed  on  the  Anseba,  the 
horns  with  a  portion  of  the  skull  of  the  ■adult  female, 
and  the  back  view  of  the  hinder  horn  of  the  latter,  all 
drawn  to  a  scale  of  one-tenth  the  natural  size.  A 
comparison  of  the  last  with  the  typical  specimen  of 
a.  keitloa  in  the  British  Museum,  or  with  the  figure 

>  Cf.  Selatar,  Studmt  and  InUU.  Obt.  1699,  p.  338. 


IIAMMAUA, 


245 


in  Dr.  Smith's  "  Illustrations  of  the  Sioology  of  South 
Africa/'  will  show  the  very  close  resemblance  in  form. 
In  the  typical  specimen  the  horns  arc  of  equal  length  ; 
in  that  from  Abyssinia  the  front  horn  is  16  in.  long, 
the  hinder  1 2f  in.  The  first  is  nearly  circular  in  section 
throughout,  being  a  little  compressed  above ;  the  latter 
is  much  compressed,  more  sharply  edged  behiiiid  than 
in  front.  Half-way  up  it  measures  2^  in.  from  front  to 
back,  and  1^  in.  across. 


The  horns  in  the  younger  specimen  measure — ^the 
anterior  12  in.  along  the  curte  in  front,  or  11  in« 
measured  in  a  direct  line  from  the  centre  of  the  base  to 
the  tip,  while  the  hiiider  horn  is  7|  in.  in  length.  The 
first  is  slightly  compressed,  having  an  oval  section; 
the  \wo  diameters  half-way  up  being  1'8  in.  and  1*4  in. 
The  hinder  horn  tapers  suddenly  below,  and  more 
gradually  above ;  it  is  rather  less  compressed  than  the 


246  ZOOLOGV. 

front  one;  the  diameters,  tliree  inches  from  the  tip,  being 
1*2  in.  and  1  in. 

In  the  young  male  in  the  Zoological  Gardens,  the 
horns  are  comparatively  short,  especially  the  hinder- 
most,  larger  at  the  base,  Uunt,  and  nearly  circular  in 
section. 

How  much  of  the  variation  in  the  three  different 
animals  is  due  to  age  and  sex,  it  is  impossible  to  say 
without  further  information  and  additional  specimens. 
It  is  quite  possible  that  the  horns  may  be  variable  in 
length,  and  in  the  extent  to  which  they  are  compressed, 
and  yet  that  R.  keitloa  may  be  perfectly  distinct 
from  R.  hicomis. 

In  both  the  animals  killed  by  our  party,  the  elongated 
process  of  the  upper  lip  was  well  developed,  as  it  is 
also  in  the  Zoological  Society's  specimen.  The  neck 
was  rather  long,  and  there  was  a  distinct  appearance 
of  a  hump  on  the  shoulder,  just  as  represented  in  Dr. 
Smith's  figure.  This  was  peculiarly  conspicuous  in  the 
larger  animal  after  death,  as  it  lay  upon  its  belly,  with 
its  legs  doubled  beneath,  a  position  taken  by  both 
animals  in  dying.  Immediately  on  returning  to  India 
after  leaving  Abyssinia,  and  referring  to  Dr.  Smith  s 
plates,  I  was  struck  by  the  great  resemblance  of  his 
figure  of  R.  keitloa,  in  the  characters  of  general  form 
as  well  as  of  the  horns,  to  the  Abyssinian  animal,  and 
I  was  glad  to  find  that  Dr.  Gray,  from  an  examination 
of  the  skull  brought  home  by  Mr.  Jesse,  aiTived  inde- 
pendently at  the  same  conclusion  as  to  the  specific 
relations  {Ann.  and  Mag.  Nat.  Hist.  1869,  p.  201). 


MAMMALIA.  247 

The  rhinoceros  of  the  Anseba  inhabits  the  dense 
thickets  on  the  bank  of  the  stream,  which  are  inter- 
sected in  all  directions  by  the  paths  made  by  these 
animals.  In  the  densest  parts,  where  roots  and  stems 
render  the  jungle  almost  impervious,  there  are  places 
known  by  the  inhabitants  as  rhinoceros-housea  The 
stems  and  branches  have  generally  been  broken  away  or 
pushed  back,  so  as  to  leave  a  clear  space  about  15  ft  or 
20  ft.  in  diameter,  at  the  bottom  of  which  the  ground 
has  been  worn  into  a  hollow  by  the  trampling  and 
rolling  of  the  animals  in  wet  weather.  These  houses 
are  used  as  retreats  during  the  heat  of  the  day.  On 
two  or  three  occasions  we  disturbed  a  rhinoceros  from 
one  of  these,  and  he  rushed  off  with  much  noise  and 
loud  snorts  through  the  bushes.  So  far  as  we  could 
learn  from  our  observations,  these  animals  enter  the 
thick  jungle  early  in  the  morning  and  rest  until  one  or 
two  o'clock  in  the  day;  then  they  leave  their  thickets 
and  go  out  to  feed,  usually  remaining,  however,  amongst 
high  bushes.  At  the  time  of  the  year  in  which  we 
visited  the  country  rain  generally  set  in  in  the  after- 
noon, and,  if  it  did  not  rain,  the  sky  was  overcast ;  in 
the  clear  weather  the  rhinoceroses  are  said  never  to 
appear  before  evening.  They  are  great  browsers,  feeding 
chiefly  on  the  young  shoots  and  branches  of  acacia 
and  other  trees,  or  on  fruits  ;  so  far  as  I  could  see,  they 
do  not  generally  eat  grass. 

Their  movements  are  very  quick,  their  usual  pace 
being  a  smart  trot,  and  the  numerous  tracks  show  that 
they  move  about  a  good  deal.     The  natives  declare  that 


248  ZOOLOGY, 

it  is  impossible  to  escape  from  them  even  on  horseback 
(I  doubt  this),  but  they  are  easily  eluded  by  turning,  as 
they  are  not  quick  of  sight,  and,  like  most  mammalia, 
they  never  look  for  enemies  in  trees ;  consequently  a  man 
only  two  or  three  feet  from  the  ground  will  remain 
unnoticed  by  them  if  he  keeps  quiet.  They  are  said 
to  be  extremely  savage,  and  unquestionably  the  first 
one  killed  by  us  charged  most  viciously.  The  same  is 
related  of  the  black  rhinoceros  by  all  African  sportsmen. 
I  cannot  help  thinking,  however,  that  their  savage  dis- 
position has  been  rather  exaggerated.  When  on  the 
Anseba,  we  heard  numerous  accounts  of  people  having 
been  killed  by  liojis,  panthers,  and  elephants ;  but  Mr. 
MUnzinger  told  me  that,  during  his  long  experience 
in  the  country,  he  never  knew  of  any  one  having 
been  killed  by  a  rhinoceros,  although  he  himself  had 
had  a  narrow  escape  when  charged  by  an  animal  he 
had  wounded. 

The  majority  of  the  animals  seen  by  our  party 
were  in  pairs,  an  old  female  and  a  nearly  full- 
grown  cub.  On  one  occasion  Captain  Mockler  saw 
four  together. 

The  ov^Y  sound  we  heard  them  make  was  the  snort 
of  alarm  or  rage^  so  frequently  referred  to  by  Sir  Samuel 
Baker  'm  his  description  of  the  rhinoceroses  met  with 
by  him  on  the  Nile  Tributaries  in  Western  Abyssinia. 
It  is  a  tnost  pecidiar  noise,  resembling  that  made  by  a 
locomotive  more  than  any  sound  made  by  an  animal 
with  which  I  am  acquainted. 

The  following  are  the  dimensions  of  the  larger  rhino- 


MjmrjLU.  249 

ceros    killed,   that  of    which   the    skeleton    is    in   the 
British  Museum : — 

ft  in. 
Length  of  head  from  snout  to  between  the  ears    ...    2  8 
„         neck  and  body  from  between  ears  to  inser- 
tion of  tail 7  1^ 

„         taU 1  9^ 

Total  length  from  nose  to  end  of  tail,  measured  along 

the  curve  of  the  back 11    7 

Length  of  body  measured  along  side  from  shoulder  to 

rump 6    9 

Height  at  shoulder ^    H 

I  am  indebted  to  Mr.  Jesse  for  these  measurementcL 
I  had,  unfortunately,  no  tape  with  me  when  we  killed 
the  animal,  and  Mr.  Jesse  took  them  from  the  carcase 
next  morning. 

14.  Hyrax  abyssinicus,  Hemp,  and  Ehr. 

H.  hahcssinicus,  Hemp,  and  Ehr.,  Symb.  Phys.,  t.  ii.,  smaller 
specimen  in  lower  figure  (nee  Euhyrax  abyssinicus,  Gray,  Ann. 
and  Mag.  Nat.  Hist.  vol.  i.  p.  47). 

The  synonymy  of  the  North  African  Hyraces  is  en- 
veloped in  the  direst  confusion.  Dr.  Gray  recently 
attempted  to  clear  this  up  in  a  paper  published  in  the 
Annals  and  Magazine  of  Natural  History  for  1868, 
Ser.  4,  vol.  i.  p.  35,  and  subsequently  reprinted  in  the 
British  Museum  Catalogue  of  "  Carnivorous,  &c.,  Mam- 
malia," p.  279  ;  but  in  consequence  of  not  having  had 
access  to  the  types  described  by  previous  authors,  he  has 
in  some  instances  mistaken  other  forms  for  them.  The 
excessive  variation  to  which   the   common  Abyssinian 


250  ZOOLOGY, 

species  is  liable  has  also  caused  it  to  be  greatly  sub- 
divided ;  but  so  much  do  skins  vary  in  colour  that  I 
am  by  no  means  surprised  that  naturalists,  having  only 
access  to  a  few  specimens  in  a  museum,  should  consider 
that  they  had  evidence  of  the  existence  of  several 
species. 

Being  aware  of  the  difficulties  which  existed,  I  en- 
deavoured, when  in  Abyssinia,  to  obtain  as  large  a  series 
as  I  possibly  could,  and  I  brought  away  with  me  twenty- 
eight  skins  from  different  parts  of  the  country.  Since 
my  return  to  Europe  I  have  not  only  compared-  these 
with  the  types  described  by  Dr.  Gray,  but  I  have  also 
examined  the  specimens  in  the  Berlin  Museum  which 
were  collected  by  Hemprich  and  Ehrenberg  and  de- 
scribed and  figured  in  the  "  Symbolae  Physicae."  The 
conclusion  at  which  I  have  arrived  is  that  there  are 
probably  five^  (perhaps  six)  distinguishable  races  of 
Hyrax  in  North-eastern  Africa,  one  or  two  of  which 
appear  to  be  new;  but  so  great  is  the  difficulty  of 
distinguishing  between  the  different  forms,  and  such  is 
the  risk  of  error,  that  I  hesitate  to  add  to  the  already 
very  considerable  list  of  names,  as  of  the  supposed  new 
species  I  have  in  one  case  one,  in  the  other  two,  spe- 
cimens only. 

The  principal  characters  relied  upon  by  Dr.  Gray  for 
the  classification  of  the  Ily races  are  the  form  of  the 
skull,  the  colour  of  the  dorsal   spot,   the  harshness  or 

^  The  two  species  not  obtained  by  me  are  H.  nificeps  vel  doiigolanus  of 
Ehrenberg,  which,  from  an  examination  of  the  types  of  both  specimens,  I 
have  ascertained  to  be  the  same  as  U,  Burtoni  of  Gray  ;  and  the  Sboa 
animal  called  Euhyrax  abyssinieus  by  Gray. 


MAMMALIA,  25 1 

softness  of  tlic  hair,  and  its  coloration.  All  of  these 
characters  appear  to  ine  to  be  Remarkably  variable  in 
the  genus  Hyrax,  and  especially  in  specimens  from 
Abyssinia. 

The  first  specimen  shot  by  me  in  Abyssinia  was  killed 
on  the  shore  of  Annesley  Bay.  It  is  quite  a  young 
animal,  and  not  fully  grown,  but  appears  smaller  than 
specimens  of  the  same  age  from  the  highlands;  the 
few  other  specimens  seen  by  me  at  the  same  time 
were  also  small.  The  skin  in  question  measures  barely 
a  foot  from  nose  to  rump.  The  sole  of  the  forefoot 
is  \\  inches  long;  that  of  the  hindfoot  1^.  The 
fur  is  unusually  thin  and  short,  being  scarcely  half 
an  inch  in  length  anywhere,  and  rather  harsh  in  texture. 
The  general  colour  is  brown ;  the  hairs  are  dull  brown 
at  the  base,  with  yellowish  tips.  There  is  a  rudi- 
mentaiy  black  dorsal  spot. 

With  some  little  doubt  I  refer  this  specimen  to  H. 
ahyssinicus.  Of  the  type  specimens  of  that  species  in 
the  Berlin  ]\Iuseura,  the  largest  is  about  eighteen  inches 
long ;  it  is  an  old  animal  with  large  teeth.  The  colour 
is  grey,  with  slight  mottling,  owing  to  the  hairs  being 
lu'own  at  the  base,  then  dirty  white  for  about  one-eighth 
of  an  inch,  and  black  at  the  tip.  The  fur  is  neither  very 
coarse  nor  very  fine.  The  under  parts  are  dull  grey.  There 
is  a  small,  not  very  distinct,  black  dorsal  spot,  the  hairs 
in  which  are  black  throughout  in  the  older  specimen,  but 
only  at  the  tip  in  a  younger  skin.  Both  specimens  are 
said  to  have  been  obtained  near  Massowa. 

The  species  identified  by  Dr.  Gray  with  Ehrenbergs 


252  ZOOLOGY. 

H.  ahyssinicus  is  a  very  distinct  form,  and,  as  stated  by 
Dr,  Gray,  the  skin  alone  is  undistinguishable  from  that 
of  Hyrax  capensis.  It  is  a  very  much  larger  species 
than  any  of  the  common  Abyssinian  forms,  very  dark- 
coloured  above  and  dusky  beneath,  with  long,  soft  fur, 
and  a  very  conspicuous  black  dorsal  spot. 

I  think  it  probable  that  the  true  Hyrax  ahyssinicus 
of  Ehrenberg  inhabits  the  neighbourhood  of  the  coast 
only.  The  highland  species,  which  is  far  better  known, 
is  the  following  : — 

15.   Hyrax  Brucei,  Gray. 

Ashkoko,  Brace's  Travels,  vol.  v.  p.  139,  with  plate. 

Hyrax  syricuiuSj  Schreber,  Saughth.  iv.  p.  314,  t.  240. 

H,  Brucei^  Gray,  Ann.   and  Mag.  N.  H.  1868,  L  p.  44.— Cat. 

Cam.  &c  Mam.  Brit.  Mns.  p.  287. 
H.  Alpini,  Cray,  Ann.  and  Mag.  1868,  p.  45. — Cat.  Brit.  Mus. 

p.  287. 
H.  ferruffineuSy  Gray,  Ann.  and  Mag.  1869,  vol.  iii.  p.  242. — Cat. 

Brit.  Mus.  p.  288. 
jff.   irrorata,  Gray,    and  var.  luteogastery  Ann.  and  Mag.  1869, 

p.  242.— Cat  Brit.  Mus.  p.  288. 

After  long  and  careful  examination  both  of  the  skins 
and  skulls  of  all  the  Abyssinian  H)n'aces  which  I  have 
been  able  to  see,  including  the  large  series  of  specimens 
obtained  by  myself,  I  am  unable  to  find  any  constant 
character  by  which  the  different  forms  above  enumerated 
can  be  discriminated.  In  several  specimens  collected 
by  me,  mostly  from  lower  elevations,  there  is  either 
no  trace  or  only  the  faintest  rudiment  of  a  dorsal  spot 
When  present  this  is  yellow.  I  was  at  first  disposed 
to  consider  those  specimens  in  which  the  dorsal  spot  is 
inconspicuous  or  absent  as  belonging  to  a  distinct  race, 


MAMMALIA,  253 

and  it  appeared  much  as  if  there  was  a  cranial  difference 
also,  the  length  of  "the  diastema  being  proportionally 
greater  in  the  unspotted  sp^imens;  but  more  careful 
examination  showed  that  it  is**piily  in  one  or  two  skins 
that  there  is  absolutely  no  trace-  of  a  dorsal  spot,  and 
that  those  belong  to  young  individuals,  in  which  it  is 
usually  less  well-marked  than  in  adults,  whilst  the 
proportional  length  of  the  diastema  is  a  character  of 
very  dubious  value.  In  two  skulls  before  me  of  about 
the  same  age,  and  of  animals  which  were,  I  believe, 
obtained  from  the  same  burrow,  the  relative  length  in 
the  upper  jaw  of  the  diastema  and  of  the  first  three 
molars  is  in  one  specimen  0'35  to  0*48  inch;  in  the 
other,  0'46  to  0*48.  The  skins  only  differ  in  one  being 
more  ferrugineous  than  the  other,  a  character  rarely  of 
much  value  amongst  mammalia  in  general,  and,  as  T  have 
ascertained  by  repeated  observations,  of  not  the  slightest 
importance  amongst  Abyssinian  Hyraces.  I  have,  in 
repeated  instances,  seen  several  animals  amongst  those 
scattered  over  the  face  of  a  cliff,  or  lying  out  on  large 
stones,  which  were  far  more  rufous  than  their  com- 
panions of  the  same  burrow. 

Of  the  specimens  collected  by  me  which  I  refer  to 
this  species,  one  was  shot  at  Upper  Sum  only  2,000  feet 
above  the  sea ;  one  at  Undul  Wells  at  about  4,000  feet ; 
three  at  Senafd,  8,000  feet ;  one  at  Agula,  one  at  Antalo, 
both  7,000  feet ;  one  in  the  Anseba  valley,  4,000  feet ; 
and  sixteen  at  Adigrat,  8,000  feet.  Amongst  this  series 
I  find  every  gradation  between  the  four  forms  described 
by  Dr.  Gray.     There  can,  I  think,  be  but  little  hesitation 


254  ZOOLOGY. 

after  examining  my  specimens  that  the  slight  differences 
of  colour  upon  which  H,  Alpini  was  distinguished  from 
H.  Brucei  are  not  of  specific  value,  and,  in  illustration  of 
this  fact,  I  may  state  that  when  collecting  in  Abyssinia 
I  never  could  quite  determine  to  which  of  the  two 
forms  I  should  refer  specimens.  As  mentioned  before, 
the  amount  of  variation  is  very  remarkable,  and  it  is  not 
surprising  that  Dr.  Gray  should  have  been  misled  by  the 
few  specimens  in  his  possession.  Hyrax  ferrugineus 
is  the  rufous  phase,  which  is  certainly  not  more  than 
an  individual  variety;  and  the  only  race  about  which 
I  ever  had  the  least  doubt  is  H.  irroratus,  which  is 
the  variety  with  the  rudimentary  dorsal  spot;  and  in 
that  case  also  I  can  see  ^o  constant  distinction  whatever. 
A  large  series  of  my  specimens  are  deposited  in  the 
British  Museum,  and  available  for  comparison,  and  I 
feel  confident  that  any  one  who  examines  them  will 
(jome  to  the  san^e  conclusion  as  myself. 

By  far  the  greatest  amount  of  yari^-tion  exists  amongst 
t'he  specimens  f^rom  Adigrat,  and  there  are  two  of  these 
which  differ  so  much  from  the  others  that  they  may 
belong  to  distinct  species.  But  after  my  experience 
pf  the  variation  of  which  Hyraces  are  susceptible,  I 
would  not  think  of  distinguishing  forms  on  the  evidence 
pf  a  single  specimen. 

The  first  of  these  aben-ant  specimens  (No.  786)  ap- 
proaches the  type  of  H.  Brucei  of  Gray ;  but  it  has 
longer  and  finer  fur,  and  is  greyer  and  less  rufous  in 
colour.  The  fur  is  dusky  at  the  base,  becoming  black 
near  the  end,  and  tipped  Nyith  white.     There  is  a  distinct 


MAMMALIA,  255 

longitudinal,  yellow,  dorsal  spot.  The  animal  is  fully 
adult,  but  not  aged.  The  skull  is  much  crushed,  and 
I  have  not  extracted  it. 

The  second  specimen  (No.  886)  differs  even  more  fcom, 
the  type,  and  very  possibly  should  be  classed  with  the 
next  species.  It  is  of  a  very  dark  brown  colour,  much 
mottled  with  black,  all  the  under  fur  being  blackish ;  the 
hairs  are  yellowish  brown  near  the  end  and  tipped  with 
black.  There  is  a  rudimentary  hlack  doprsal  spot,  the 
hairs  on  the  centre  of  the  back  being  quite  black  near 
the  base,  and  only  very  slightly  tipped  with  pal^r  colour. 

The  habits  of  the  Abyssinian  Hyraces  are  precisely 
similar  to  those  of  Syria  and  the  Cape.  They  live  in 
rocky  or  stony  places,  in  communities,  like  rabbits, 
haunting  holes  beneath  the  rocks.  A  large  pUe  of  loose 
blocks,  especially  if  there  ^re  precipices  around,  is  sure  to 
be  inhabited  by  them.  They  are  frequently  found,  too, 
in  rocky  watercourses.  They  appear  to  feed  at  night 
and  very  early  in  the  morning,  their  principal  food  being 
the  leaves  and  young  shoots  of  trees  and  bushes.  In, 
the  stomach  I  invariably  found  a  green  mass  too  much 
crushed  for  the  separate  leaves  to  be  distinguished. 
During  the  day  they  lie  out  upon  rocks  in  the  shade,^ 
or  retire,  especially  towards  midday,  beneath  the  rocks. 
They  ai-e  timid  and  wary,  rushing  into  their  holes  at 
the  smallest  intimation  of  danger.  The  only  sound  \ 
heard  made  by  them  was  a  shrill  squeak  when  suddenly- 
alarmed.  They  can  climb  over  smooth  surfaces  of  rocl^ 
in  a  wonderful  manner,  their  large  flat  feet  aiding  theii\ 
in  obtaining  a  hold. 


256  ZOOLOGY, 

The  following  measurements  were  taken  from  a  freshly- 
killed  adult  female  shot  at  Senafd  : — 

ft.  in. 

LeDgth  of  face,  from  end  of  nose  to  between  ears   .    .    0  3*5 

„       of  neck,  from  between  ears  to  top  of  shoulder     0  3 

„       from  shoulder  to  rump 1  1*5 

Total  length,  measured  along  the  curve  of  the  back     .     1     8 

Length  of  body,  from  shoulder  to  rump,  niensured 

along  the  side 13 

Height  at  shoulder 0  8 

Girth  of  neck 0  7*6 

„      chest  behind  the  shoulder 1  1*5 

Length  of  ear 0  0*75 

Breadth  of  ear 0  0*87 

Length  of  longest  hairs  in  moustache 0  2*5 

„        forefoot  and  toes 0  1*62 

„        middle  toe  of  forefoot 0  *3 

„        radius 0  2*5 

„        hindfoot  and  toes 0  1*62 

„        outer  toe  of  hindfoot 0  *3 

„        middle  toe  of  hindfoot 0  *G 

„        inner  toe  of  hindfoot 0  *75 

„        tibia 0  3*5 

There  are  six  mammse,  four  inguinal  and  two  pectoral : 
the  former  are  very  lateral,  just  in  front  of  the  thigh 
and  about  f -inch  apart. 

In  one  of  the  skulls  brought  back  by  me  belonging 
to  a  very  aged  animal  killed  at  Sum,  there  is  a  very 
remarkable  dental  character.  At  the  back  of  all  the 
molars  on  each  side  of  the  upper  jaw  there  is  a  pointed 
tusk-like  tooth,  shaped  like  a  canine,  the  point  projecting 
a  little  above  the  crowns  of  the  molars.  The  series  of 
grinders  is  complete,  being  seven  in  number,  and  the 
teeth  are  considerably  worn.  The  canines  in  this  speci- 
men are  rounded  in  front,  not  ridged  afl  usual 


MJMMJLU,  257 

Of  the  three  skeletons  brought  back  two  have  twenty- 
one,  one  twenty  pairs  of  ribs.  All  have  twenty-eight 
dorsal  vertebrae. 


x6.  Myrax,  tsp.  nov. 

Two  specimens  fix)m  Gaso  and  Santara  on  the  Dalanta 
plateau  differ  from  aU  the  others  in  having  a  distinct 
small  black  dorsal  spot  They  are  very  dark  and  much  less 
mottled  than  usual ;  the  hair  is  longer  and  moderately 
fina  The  soles  of  the  feet,  of  the  hinder  ones  espe- 
cially, appear  to  be  rather  shorter  than  usual  The 
nasal  bones  of  the  skull  appear  shorter,  but  neither 
specimen  is  quite  adult.  Still  compared  with  specimens 
of  H.  Brucei  of  the  same  age  there  appears  a  diflference, 
the  height  of  the  posterior  portion  of  the  skull  in  the 
latter  being  less  in  proportion  to  the  length.  In  the 
Dalanta  specimens  the  zygomatic  arch  is  broader  than 
usual.  The  series  of  molars  in  the  upper  jaw  are  much 
curved. 

I  am  inclined  to  consider  this  a  distinct  form,  and  I 
doubt  if  it  be  the  young  of  the  Shoa  species  called 
Euhyrax  ahyssinicus  by  Dr.  Gray.  It  appears  to  me 
a  smaller  animal  My  largest  specimen  is  nearly  adult 
and  wants  the  small  front  premolar  on  each  side  of  both 
jaws.  This  tooth  is  occasionally  wanting  here  and  there  in 
other  skulls,  and  is  generally  deficient  in  the  lower  jaw 
of  aged  specimens,  but  amongst  eight  adult  skulls  before 
me  I  find  no  other  instance  of  its  absence  throughout 
both  jaws. 

5{ 


258  ZOOLOGY. 

17.  Elephas  (Loxodonta)  Africanus,  Cuvier. 
CuY.  B^ne  Animal,  ii.  p.  231. 

Elephants  were  not  very  uncommon  in  the  neighbour- 
hood of  Annedey  Bay,  and  several  were  killed  by  some 
of  the  sportsmen  in  the  army,  whilst  their  footprints 
and  the  dried  droppings  were  conspicuous  everywhere. 
Farther  north,  near  Massowa,  we  met  with  their  tracks 
in  all  the  valleys  traversed,  but  we  very  rarely  saw  the 
animals  themselves,  and  we  only  killed  a  small  herd  of 
five  at  Kokai,  in  the  Lebka  valley,  as  related  on  a 
previous  page. 

The  diflferences  between  the  form  and  habits  of  the 
Indian  and  African  Elephants  have  been  so  frequently 
described,  and  both  species  have  of  late  years  become 
so  well  known  by  the  exhibition  of  specimens  in  the 
various  Zoological  Gardens  in  Europe,  that  but  few  re- 
marks are  necessary.  The  impression  produced  upon  all 
who  saw  the  Abyssinian  Elephants  was  that  they  were 
more  active  and  better  climbers  than  the  Indian  animals, 
and  also  that  they  were  more  savage  in  disposition. 
They  often  inhabit  steep  hill-sides,  and  the  rapidity  with 
which  they  ascend  and  descend  these  was  described  as 
marvellous  by  some  of  those  who  witnessed  them. 

The  conduct  of  the  small  herd  at  Kokai  was  very 
curious,  illustrating  the  boldness  of  the  creatures  in 
remaining  all  day  in  the  midst  of  men  and  domestic 
animals,  while  at  the  same  time  they  afforded  a  re- 
markable case  of  stupidity  when  attacked.  On  a 
subsequent  occasion  Captain  Mockler  attacked  a  herd 
of  very  large  Elephants ;  and  at  the  first  shot  the  whole 


MAMMALIA.  2S9 

herd  turned  upon  him  although  they  were  previoualy 
quite  unaware  of  his  presence,  and  he  only  saved  his 
life  by  running  and  hiding  himself  in  a  bush. 

The  Elephants  near  the  Abyssinian  coast  migrate  with 
the  season,  coming  to  the  coast  in  the  winter  and  spring, 
when  rain  falls  and  there  is  abundance  of  green  food, 
and  returning  to  the  hills  about  June  and  July,  when 
the  monsoon  rain  commences.  They  ascend  the  various 
plateaux  near  the  Anseba  valley  to  a  height  of  7,000  or 
8,000  feet,  partly  in  all  probability  to  escape  from  the 
flies,  which  are  very  troublesome  in  the  lower  ground  at 
that  season. 

All  of  the  Elephants  in  Eastern  and  Northern  Abyssinia 
appear  to  be  almost  tuskless  or  to  have  very  small  and 
short  tusks,  a  most  unusual  occurrence  in  the  African 
Elephant,  though  tuskless  varieties  of  the  Asiatic  species 
are  extremely  common  in  India  and  other  parts  of  South- 
eastern Asia,  and  in  Ceylon  a  tusker  is  very  rare  indeed. 
In  Western  Abyssinia,  on  the  Barka,  Mareb,  and  Atbara, 
the  Elephants  have  fine  tusks,  and  are  much  hunted  in 
consequence. 

Sir  Emerson  Tennent,  in  his  description  of  the  Elephant 
of  Ceylon,  especially  notices  the  circumstance  that  all 
the  animals  in  one  herd  are  generally,  if  not  always,  part 
of  the  same  family.  The  little  herd  killed  by  us  at 
Kokai  evidently  consisted  of  an  old  female  and  four 
younger  Elephants,  her  offspring  in  different  stages  of 
growth ;  and  subsequently  Captain  Mockler  saw  a  herd 
of  nine,  consisting  in  a  similar  manner  of  a  very  large 
femaler  and  of  eight  younger  animals  varying  in  size 

S2 


260  ZOOLOGY, 

from  adult  down  to  a  little  beast  not  higher  than  a 
Shetland  pony.  It  is  very  probable  that  all  herds  consist 
of  animals  closely  related  by  consanguinity. 

The  dimensions  of  the  old  female  and  of  the  largest 
male  killed  at  Kokai  were : — 

FratAi/E.         Male. 
ft    in.         ft    in. 

Height  at  shoulder  ..........    7    8  74 

From  top  of  head,  between  base  of  ears  to  top 

of  shoulder 15  16 

From  top  of  shoulder  to  insertion  of  tail    ..73  63 

Length  of  body  from  shoulder  to  rump,  mea- 
sured along  the  aide  in  a  direct  line     ..73  66 

Circumference  of  fprefoot  3    7^  — 

Obpbr  HUMINANTIA. 

z8.  Gazella  Scemmerringii,  EUpp. 

AnXHope  Sotmmerrififfii,   Riipp.  Atlas,  t.   19. — Brehm,  Habesch, 
p.  65. 

This  Antelope  abounds  on  the  coast  of  the  Red  Sea, 
near  Annesley  Bay  and  Massowa,  but  never  ascends  the 
hills.  It  inhabits  principally  the  low  bush  and  acacia 
scrub,  and  is  found  in  herds  varying  fix)m  a  few  indi- 
viduals up  to  100  or  more.  It  appeared  to  me  that  the 
herds  were  larger  in  the  winter  than  in  the  summer,  but 
around  Annesley  Bay  many  had  been  driven  from  their 
usual  haunts  by  the  imwonted  presence  of  so  many  men, 
and  several  herds  must  frequently  have  united.  In  one 
instance,  at  Arafild,  an  immense  assemblage  of  at  least 
300  to  400  inhabited  a  large  plain  south  of  the  Egyptian 
camp.  A  very  large  number  were  shot  by  the  sportsmen 
of  the  army.     The  flesh  is  excellent 

Soemmerring's  Gazelle  belongs  to  the  same  section  of 


1.  CAZELLA     DOHCA: 


G     AHABICA. 
».        »    C.'SUBCUTTUROSA. 
BENNETTI.       S    C     SPEKEI 


MAMMALIA.  261 

the  genus  as  G.  dama  and  O.  euchore^  distinguished 
from  G.  dorcas  and  its  allies,  the  true  gazelles,  by  their 
less  symmetrical  form  and  longer  legs,  and  by  their  habit 
of  keeping  in  herds  of  considerable  size.  Their  pace  is 
rapid,  usually  a  long  trot :  I  never  saw  them  bound  like 
the  Springbok  or  like  Antilope  bezoartica  of  India. 

On  one  or  two  occasions  I  saw  tracks  of  Gazella 
Sosmmerringii  near  water,  and  Captain  Mockler  shot 
one  while  drinking.  The  hour  of  drinking  appears  to 
be  usually  a  little  before  midday.  In  this  it  diffisrs 
from  Gazella  dorcas  and  its  near  ally  G.  Bennetti  of 
India,  which  never  drink. 

19.  G.  dorcas,  L. 

(hpra  d&rcoi,  L.  Syst.  Nat,  Ed.  ISyi,  L  p.  96. 

The  true  Gazelle  of  the  Abyssinian  coast-land  appears 
to  diflfer  in  no  essential  character  from  that  of  Northern 
Africa.  A  distinguishable  race  inhabits  the  opposite 
shore  of  the  Red  Sea  {G.  ardbica^  H.  and  E.),  and  a 
third  form,  G.  Spekei,  Blyth,  occurs  on  the  African  coast 
further  south  in  the  Somali  country.  Figures  of  the  horns 
of  all  these  races  are  appended  (Plate  I),  as  well  as  of 
two  other  Asiatic  Gazelles,  G.  suhgutturosa  of  Persia  and 
Beloochistan,  and  G.  Bennetti  of  India.  The  figures 
of  G.  dorcas  are  from  specimens  shot  by  myself  near 
ZuUa ;  that  of  G.  arabica  from  a  head  for  which  I 
am  indebted  to  Captain  Heysham  of  the  Commissariat, 
who  obtained  it  at  Mocha.  The  figured  heads,  male 
and  female,  of  G.  Spekei,  are  the  specimens  described 
by  Mr.  Blyth  in  the  Journal  of  the  Asiatic  Society 


262  ZOOLOGY. 

of  Bengal  for  1856,  vol.  xxiv.  p.  296,  and  now  in 
the  IndiiOn  Museum  at  Calcutta.  Those  of  G.  suhgut- 
turosa  are  in  the  same  museum,  and  were  brought 
from  Afghanistan,  I  believe,  by  Captain  Hutton.  The 
figures  of  the  homs  of  G.  Bennetti  are  taken  from  an 
animal  shot  by  myself  in  Western  India.  The  length 
of  the  homs  varies  much  with  age,  but  the  curve 
appears  to  be  fairly  constant. 

So  far  as  my  observation  extends,  neither  the  Dorcas 
nor  Bennett's  Gazelle  are  ever  seen  in  large  flocks,  like 
the  animals  of  the  Springbok  group.  Usually  both  are 
seen  solitary  or  from  two  to  five  together,  inhabiting 
thin  bushes,  generally  in  broken  ground.  They  feed  much 
upon  the  leaves  of  bushes.  The  male  has  a  peculiar 
habit  when  surprised  of  standing  still  and  uttering  a 
short  sharp  cry.  Like  most  Antelopes,  they  keep  much 
to  the  neighbourhood  of  some  particular  spot.  After 
long  observation,  I  am  convinced  that  Bennett's  Gazelle 
never  drinks,  and  all  that  I  could  ascertain  of  the  Dorcas 
Gazelle  leads  to  the  same  conclusion  in  its  case.  It  would 
be  interesting  to  ascertain  if  all  these  peculiarities  are 
shared  by  the  little  group  allied  to  G.  dorcas,  or  the 
true  GazeUes,  as  distinguished  from  the  group  to  which 
belong  G.  dama,  G.  eit^hore,  and  G.  Scemmerringii. 

ao.  Oryx  beisa,  Riipp. 

AntUope  heiaOf  Biipp.  Netu  Wirb.  p.  14,  t.  S.^Brehm,  Habesch, 
p.  66. 

I  never  saw  any  other  Antelope  except  the  two  species 
of  Gazelle  and  Neotragus  Hemprichii  near  Annesley 
Bay,  and  I  never  heard  of  the  occurrence  of  Oryx  till 


MAMMALIA.  263 

I  was  north  of  Massowa.  On  our  journey  to  the  Anseba 
and  Bogos,  we  met  with  none,  though  Captain  Mockler 
and  I  found  their  fresh  tracks  near  one  of  the  drinking- 
places ;  but  when  returning,  I  stopped  for  three  days  at 
a  halting-place  in  the  semi-desert  north  of  Massowa,  and 
I  succeeded  in  shooting  four  of  these  superb  and  rare 
Antelopes.  All  were  females:  but  there  is  little,  if 
any,  diflfercnce  in  the  sexes,  both  having  equally  fine 
horns. 

The  Beisa  is  found  singly  or  in  small  herds,  rarely 
exceeding  ten  in  number,  in  the  somewhat  hilly  barren 
country  near  the  sea-coast.  They  are  said  to  keep  to 
the  more  hilly  parts  of  Samhar.  Near  Annesley  Bay, 
where  the  country  is  more  wooded,  this  Antelope  does 
not  occur,  but  it  abounds  farther  south  in  thb  Somali 
country,  and  the  horns  are  brought  in  considerable 
numbers  to  Aden  from  Berbera.  They  are  used  as 
weapons  by  the  Somalis. 

The  principal  food  of  the  Oryx  near  Massowa  is  a 
coarse  grass  almost  resembling  a  diminutive  bamboo. 
They  appear  to  be  grazers  rather  than  browsers,  although, 
like  all  Antelopes,  they  occasionally  eat  the  young  shoots 
of  Acacia  and  other  trees.  They  are  quite  diurnal  in 
their  habits,  feeding  in  the  morning  and  evening,  in 
this  respect  resembling  the  Gazelles,  to  which  they  are 
unquestionably  closely  allied.  When  we  were  in  the 
Samhar  country  in  July  and  August,  the  Oryx  drank 
apparently  every  day,  always  coming  to  the  water  about 
one  or  two  o'clock.  It  is  probable  that  they  drink  less 
regularly  in  cold  weather. 


264  ZOOLOGY, 

The  appearance  of  a  herd  of  Oryx  is  very  imposing. 
They  are  some  of  the  most  elegant  and  symmetrical  of 
animals,  the  motions  being  those  of  a  wild  horse  rather 
than  of  an  Antelope.  Their  favourite  pace  appears  to  be 
either  a  steady  quick  wlak  or  a  trot ;  they  rarely  break 
into  a  gallop,  unless  greatly  alarmed.  When  frightened, 
they  dash  oflf,  sometimes  snorting  and  putting  their 
heads  down  as  if  charging,  raising  their  long  tails,  and 
looking  very  formidable.  They  are  wary  animals,  though 
far  less  so  than  some  other  Antelopes.  It  is  said  that 
they  frequently  attack  when  wounded,  and  their  long 
straight  horns  are  most  deadly  weapons.  Stories  arc 
told  of  the  Cape  Oryx,  an  allied  but  distinct  species, 
killing  Lions ;  and  I  heard  of  a  commissariat  contractor 
who  was  bringing  Camels  to  Annesley  Bay,  and  wounded 
an  Oryx  near  Massowa,  being  charged  by  it :  but  nothing 
of  the  kind  happened  to  myself,  although  I  had  to 
pursue  two  of  the  animals  I  killed  for  a  long  distance 
after  wounding  them ;  and  one  with  a  broken  leg  I  fairly 
ran  down.  The  shooting  of  these  animals  is  described 
on  an  earlier  page. 

Like  the  Gazelles  and  true  Antelopes,  all  equally 
inhabitants  of  deserts  and  open  plains,  the  Oryx  has  a 
pointed  foot,  each  of  the  divisions  being  rudely  tri- 
angular.  Its  tracks  may  consequently  be  instantly 
distinguished  from  those  of  cattle  or  of  any  of  tho 
bovine  Antelopes.  So  far  as  my  acquaintance  with  tho 
family  goes,  most  of  the  forest  and  bush-haunting  Ante- 
lopes, Koodoo — Nylgai,  Tetraceros — have  their  feet 
formed  like  those  of  the  Cervidw,  with  rounded  hoofs, 


MAMMALIA.  265 

whilst  the  Antelopes  of  the  plain,  and  especially  desert 
forms,  have  pointed  hoofs. 

The  following  are  the  dimensions  of  a  female  : — 

ft    In. 

Length  of  fieuM  from  nose  to  between  ears 15 

,,  neck  from  between  ears  to  top  of  shoulder  .  1  6 
„  back  from  top  of  shoulder  to  insertion  of  tail .  3  1 
„       tail,  1  ft  3  in.,  hairs  beyond  end  11  in.,  total.    2    2 

Total  length  from  nose  to  end  of  tail  measured  along 

the  curve  of  the  back 8    2 

Length  of  the  body  measured  in  a  direct  line  along 

the  side,  from  the  front  of  the  shoulder  to  the  rump  3    7 

Height  at  shoulder 3    7 

Girth  of  chest  behind  shoulders 3  10 

„      neck 18 

„      head  over  forehead  ..........  2    0 

„      forearm 0    9 

Length  of  horns 2    6 

„      ear 4    8 

Width  of  ear 0    4 

Xiength  of  forearm  (measured  outside  skin) 12 

„      cannon  bone  „  0    9 

„      fetlock  totoe 36 

„      thigh  from  stifle  to  hock,  measured  in  front, 

outside  skin 12 

„      back  to  hind  fetlock,  also  measured  in  front  1    2 

„      fetlock  to  hind  toe 06 

21.  Oreotragus  saltatrix  (Bodd.). 

Antilope  Oreotragus,  Gmel.  Syst.  Nat.  i.  p.  189. 
Oreotragus  saJtatrix,  Brehm,  Habesch,  p.  65. 

The  Klipspringer  is  common  on  the  more  rocky  of  the 
Abyssinian  hills,  from  a  height  of  about  3,000  feet  above 
the  sea,  or  rather  less,  to  8,000  or  9,000.  In  the  pass 
below  Senafd,  and  in  that  leading  from  Ain  to  the  An- 
seba,  by  the  valley  of  the  Lebka,  these  little  mountain 


266  ZOOLOGY. 

Antelopes  were  frequently  seen,  and  they  were  common 
on  some  of  the  rocky  precipices  on  the  flanks  of  the  great 
valleys  around  Senaf^  Guna  Guna,  Fokada,  &c.,  usually 
solitary  or  in  pairs.  When  alarmed  they  frequently  perch 
on  the  very  highest  rocks,  their  agility  in  leaping  from 
crag  to  crag  being  remarkable. 

The  very  peculiar  face  of  the  Klipspringer  resembles  in 
a  most  remarkable  manner  that  of  the  Musk-deer  of  the 
Himalaya  and  Central  Asia,  Moschtcs  moschiferus,  and 
the  general  build  is  also  similar.  There  is  probably 
much  resemblance  in  their  habits,  but  the  Musk  Deer 
appears  to  be  more  of  a  forest  animal.  The  following 
are  the  dimensions  of  a  female  Klipspringer  shot  at 
Senafd : — 

ft    in. 

Length  of  face  from  nose  to  between  ears 0  6 

„        neck  and  body  from  between  ears  to  insertion 

of  taa 2  3 

„        tail,  indading  haiis  at  end 0  4 

Total  length  from  nose  to  end  of  tail,  measured  along 

the  canre  of  the  back 3    1 

Length  of  the  body  measured  in  a  direct  line  along 
the  side,  from  the  front  of  the  shoulder  to  the 

rump 1  11 

Height  at  shoulder 16 

Girth  of  chest  behind  shoulders 1  10 

„     neck 10 

Length  of  ear 0    4 

aa.  Scopophorus  montanus,  Eiipp. 
AfUUope  marUana,  Riipp.  Atlas,  t  3. 

This  peculiar  little  Antelope  was  rare  in  the  country 
traversed ;  I  only  saw  it  two  or  three  times  near  Dolo  and 


MJMMAUA.  267 

Harkhallat,  north  of  Antalo,  at  an  elevation  of  about 
7,000  feet  above  the  sea.  It  inhabits  bushy  ground  or 
high  grasa 

I  shot  a  buck,  but  omitted  to  take  the  complete  dimen- 
sions ;  it  was  1  ft.  10^  in.  high  at  the  shoulder,  and  the 
length  of  the  body  measured  in  a  straight  line  from  the 
shoulder  to  the  rump  was  26  inches.  The  mammae  are 
four  in  number,  the  suborbital  and  inguinal  glands  well 
developed,  and  it  has  the  peculiar  bare  spot  at  the  base 
of  each  ear  characteristic  of  this  genus  of  Antelopes. 

23.  Cephalophus  madoqua,  Eiipp. 

AntUope  madoqua,  Riipp.  Neu.  Wirb.  p.  22,  t  7,  p.  2. 

This  bush  Antelope  is  not  rare  on  the  highlands  at 
elevations  of  7,000  to  8,000  feet.  I  usually  saw  it  soli- 
tary.    It  keeps  to  bushes  and  small  clearings. 

I  shot  two  bucks ;  one  at  Dildi  (1),  the  other  at  Don- 
golo  (2).    The  following  are  the  dimensions  : — 

1.  2. 

ft    in.         ft.   liL 

Length  of  face  from  nose  to  base  of  horns   ..06  06 

„        neck  from  base  of  horns  to  top  of 

shoulder 08^  09 

„  back,  from  top  of  shoulder  to  inser- 
tion of  tail      17  1     7^ 

„        tail 0    4  0    4i 

„        hairs  at  end  of  tail 0    3  0    3 

Total  length  from  nose  to  end  of  tail  measmed 

along  the  curve  of  the  back 3    4j^      3    6 

s 

Length  of  body,  measured  in  a  straight  line 
along  the  side,  from  front  of  shoulder  to 
rump 1  10        1    9i 

Height  at  shoulder 19        1  10 

„        rump —  1  11  - 


in. 

ft 

in. 

8J 

1 

9i 

9 

0 

9i 

oi 

1 

1 

6 

0 

6i 

4J 

0 

4 

2 

0 

2 

0 

8 

» 

0 

7 

0 

2i 

268  ZOOLOGY, 

1 

ft. 

Girth  of  chest  behind  shoulder 1 

,)     neck 0 

Length  of  foreleg 1 

of  radius 0 

from  knee  to  fetlock 0 

from  fetlock  to  heel  of  hoof   ....  0 

of  thigh  (tibia) — 

from  hock  to  hind  fetlock — 

from  fetlock  to  heel — 

The  suborbital  gland  is  closed     It  is  brown. 
94.  Neotragus  saltianus,  Blainv. 

AntUope  '  saltianay  Blainy.    BulL   Soc.  Philom.    1816.~Biippell, 

Atlas,  t.  21. — Hempr.  and  Ehr.,  Sjmb.  Phys.,  t.  viL 
AntUope  hemprichianay  Uempr.  and  Ehr.,  Symb.  Phys. 

The  "  Beni  Israel "  or  "  Om-dig-dig"  one  of  the 
smallest  Antelopes  known,  abounds  on  the  shores  of  the 
Bed  Sea  and  throughout  the  tropical  and  sub-tropical 
regions  of  Abyssinia.  It  is  occasionally,  but  rarely,  found 
at  higher  elevations;  I  heard  of  instances  of  its  being 
shot  both  at  Senaf^  and  Dildi ;  but  it  is  not  often  seen 
above  about  6,000  feet.  It  inhabits  bushes,  keeping 
much  to  heavy  jimgle  on  the  banks  of  watercourses,  and 
is  usually  single,  or  in  pairs,  either  a  male  and  female  or 
a  female  and  young  being  found  together ;  less  often  the 
female  is  accompanied  by  two  younger  ones,  which 
remain  with  her  until  full-grown. 

Brehm  gives  a  long  and  most  interesting  account  of 
the  habits  of  this  little  Antelope,  which  appears  to  me  to 
agree  generally  with  my  own  observations.  The  fact  to 
which  he  particularly  calls  attention,  and  which  he  gives 
on  the  authority  of  a  missionary  who  had  long  resided  at 


MAMMALIA.  269 

Keren,  is  the  attachment  which  these  animals  show  for 
particular  spots,  in  which  they  may  be  found  day  after 
day  ;  and  yet  if  the  animal  which  occupies  a  particular 
station  be  killed,  another  will  be  found  in  many  instances 
within  a  few  days  to  have  taken  possession  of  the  vacant 
post.  I  had  no  opportunity  of  noticing  whether  this  was 
the  case  with  the  Beni  Israel,  but  it  is  so  common  an 
occurrence  amongst  wild  mammals,  not  merely  with 
Ruminants,  but  with  Carnivora  also,  that  I  have  no  doubt 
of  its  accuracy.     It  is  not  even  peculiar  to  mammalia. 

Like  Gazella  dorcas  and  m^y  larger  Antelopes,  the 
Beni  Israel  has  the  habit  of  depositing  its  dung  fre- 
quently on  the  same  spot,  so  that  its  usual  haunts  may 
be  known  by  little  piles  of  its  droppings.  It  rarely 
leaves  the  shelter  of  the  bushes  during  the  day,  and  is,  I 
suspect,  somewhat  nocturnal  in  its  habits,  as  I  have  seen 
it  feeding  on  leaves  at  the  edges  of  the  jungle  in  the  dusk 
of  evening. 

All  the  specimens  of  Salt's  Antelope  seen  in  the  Anseba 
valley  differed  from  those  of  the  coast  and  of  the  pass 
between  Komayli  and  Senaf6  in  their  much  more  rufous 
colour.  There  being  no  distinction,  so  far  as  I  can  see, 
in  size  or  shape,  I  am  inclined  to  look  upon  this  as  an 
unimportant  variation — the  more  so  that,  as  previously 
noticed  when  speaking  of  the  Hyraces,  many  animals, 
and  especially  mammals,  have  a  tendency  at  times  or  in 
particular  localities  to  assume  a  rufous  phase ;  so  that  the 
difference  between  rufous  and  grey,  or  rufous  and  brown, 
is  one  of  the  least  characteristic  and  certain  of  specific 
distinctions. 


270  zooLoor. 

The  following  are  the  dimensions  of  a  full-grown 
female  of  Salt's  Antelope  (the  specimen  measured  being 
the  rufous  variety) : — 

ft    in. 

Length  of  &ce  from  nose  to  between  ears 0  5 

neck  from  between  ears  to  top  of  shoulder    .  0  7 

back  from  top  of  shoulder  to  rump   ....  1  1^ 

tail,  1^  in. ;  hairs  beyond  end  of  tail,  1  in.    .  0  2i^ 


Total  length  from  nose  to  end  of  tail,  measured  along 

the  curve  of  the  back  2    4 

Length  of  the  body,  measured  in  a  straight  line  from 

the  front  of  the  shoulder  to  the  rump 13 

Height  at  shoulder , 14 

Girth  of  chest  behind  shoulder 13^ 

„         neck 0    5^ 

„         head  in  front  of  the  ears 0    6^ 

„         forearm 0    3 

Length  of  ear 0    3^ 

Longest  hairs  of  forehead  tuft 0    1| 

Length  of  foreann  (radius),  from  elbow  to  knee   ...  0    4^ 

„      of  cannon-bone  from  knee  to  fetlock    ....  0    3^ 

„      from  fetlock  to  toe 0    2^ 

„      of  thigh  (tibia)  from  hip  to  hock 0    5| 

„      from  hock  to  hind  fetlock 0    4^ 

„      from  fetlock  totoe 02^ 

The  flesh  of  Salt's  Antelope  is  inferior  to  that  of  any 
others  that  I  have  ever  eaten ;  not  only  is  it  hard  and 
dry,  but  it  has  a  peculiar  unpleasant  musky  flavour. 

25.  Strepsiceros  Kudu,  H.  Smith. 

AniUopt  itr^piieeroif  GmeL  Syst  Nat  I  p.  192. 

The  Koodoo  is  by  £eup  the  finest  of  the  Antelopes  inhabit- 
ing the  Abyssinian  highlands.  It  occurs  in  all  the  wilder 
and  more  wooded  parts  of  the  country  from  an  elevation 
of  about  3,000  feet  to  8,000  or  9,000,  or  throughout  the 
sub-tropical  and  temperate  zones.    It  is  usually  found  in 


MAMMALIA.  271 

forest  or  bush  jungle,  rarely  in  the  open,  except  in  the 
early  morning  and  evening,  and  it  appears  to  love  the 
hill-sides  rather  than  level  country.  The  males  are 
generally  seen  solitary  or  two  or  three  together;  the 
females  and  young  animals  in  herds  of  from  four 
or  five  to  about  fifteen.  At  those  periods  of  the  year 
in  which  I  most  frequently  saw  Koodoo,  viz.  around 
Senaf(6  in  February  and  March,  and  in  the  Anseba  valley 
in  July  and  August,  the  bucks  were  rarely  met  with  in 
company  of  the  does. 

The  Koodoo  is  a  very  timid  and  wary  animal,  probably 
partly  on  account  of  the  persecution  it  undergoes.  Many 
are  killed  by  the  Abyssinians,  who  are  not,  however,  great 
hunters ;  perhaps  even  a  larger  number  fall  a  prey  to 
Lions.  It  is  probable  that  this  magnificent  Antelope 
might  be  acclimatized  in  Europe,  where  it  would  certainly 
be  a  valuable  acquisition.  The  flesh  is  delicious,  and  the 
animal  is  very  handsome,  the  large  spiral  horns  of  the 
male  being  scarcely  equalled  in  beauty  by  any  other  of 
the  family.  The  bluish  grey  colour  and  white  stripes 
are  peculiar.  As  usual,  the  adult  male  is  darker  and 
less  rufous  than  the  female,  and  the  young  is  much 
redder,  being  almost  fawn-coloured. 

The  following  are  the  dimensions  of  an  adult  but  not 
aged  male,  shot  in  the  Anseba  valley : — 

ft    in. 

Length  of  face  from  nose  to  base  of  homs 12 

„       from  base  of  homs  to  top  of  shoulder  ....  2    6 

„       from  top  of  shoulder  to  insertion  of  tail   ...  3    8 

„       of  tail,  including  hairs  at  the  end 16 

Total  length  from  nose  to  end  of  tail,  measured  along 

the  cunre  of  the  back 8  10 


272  ZOOLOGY. 

(t.    in. 
Length  of  the  body,  measured  in  a  direct  line  along 
the  side,  from  the  front  of  the  shoulder  to  thd 

rump 4    5 

Height  at  shoulder 4    4 

Length  of  forearm  (radius)  from  elbow  to  knee     ...14' 

,,      of  cannon-bone  from  knee  to  fetlock     ....  I     1^ 

„      of  thigh  (tibia) 14 

„      from  hock  to  hind  fetlock  ........  1     4 

„      of  each  horn,  measured  in  a  straight  line  from 

point  to  base 2    S 

„      of  each  horn,  measured  along  curve      ....  2  10 

Distance  between  points  of  horns 2    3 

Length  of  eara 10 

The  horns  were  only  of  moderate  size  ;  in  large  specimens 
they  are  nearly  three  feet  long. 

.  Mr.  Blyth  has  lately  shown  (P.  Z.  S.  1869,  p.  51)  that 
a  smaller  race  of  Koodoo  exists  in  S.  Africa,  and  probably 
in  Abyssinia.    I  did  not  meet  with  this  race  or  its  horns. 

Domestic  Ruminants. 

Before  quitting  the  subject  of  the  Ruminantia,  a  few 
words  on  the  domestic  cattle,  sheep,  and  goats  may  not 
be  out  of  place. 

AJl  the  oxen  of  Abyssinia  belong  to  the  humped 
race,  Bos  zehuy  and  have  the  peculiar  voice  and  habits 
of  their  species.  The  hump,  however,  is  smaller  than 
in  the  cattle  of  Hindustan ;  I  never  saw  bulls  with  the 
very  remarkable  dorsal  boss  so  characteristic  of  the 
better-bred  Indian  animals.  The  Abyssinian  cattle  are 
small  in  general :  the  finest  seen  by  me  were  those  of 
the  Wadela  and  Dalanta  plateaux.  They  vary  in  colour, 
being  mostly  brown  or  brown  and  white.  The  horns 
are  usually  ill-shaped.     The  famous  Galla  oxen  appear 


MAMMALIA,  273 

to  be  peculiar  to  the  countries  lying  east  of  Antalo,  and 
I  saw  none  myself. 

The  sheep  kept  by  the  Abyssinians  of  the  highlands 
belong  entirely,  so  far  as  my  observations  extended,  to 
the  short  and  fat-tailed  race  {Ovis  steatopygus).  They 
are  covered  with  wool,  not  like  the  well-known  Somali 
breed,  commonly  known  throughout  the  East  as  Aden 
sheep,  which  have  a  clothing  of  hair.  The  horns  are 
frequently  well-developed  and  handsomely  curved.  The 
sheep  kept  by  the  Habab  tribes  near  the  coast,  and  by 
the  Shohos,  belong  to  the  long-tailed  race,  and  much 
resemble  the  Indian  sheep.  They  have  short  wool,  and 
are  generally  hornless. 

Immense  numbers  of  goats  are  kept  by  the  Shohos 
and  Habab  tribes.  The  males  are  very  fine  animals, 
with  horns  sometimes  two  feet  in  length,  twisted  slightly, 
and  very  upright,  much  resembling  those  of  the  Markhor 
(Capra  megaceros). 

Order  RODENTIA, 

26.  Lepus  segyptius,  Geoffroy. 

Description  de  P  Egypte. 

L.  agifptiacus,  Hemp,  and  Elir.,  Symb.  Phys.,  t.  xv.  fig.  1. 

L.  lihycus,  H.  and  E.  (Berlin  Museum). 

L.  hahessinicus,  Brehm,  Habesch,  p.  64. 

Z.  abyssinicus,  Gray,  Ann.  and  Mag.  Nat.  Hist.  1867,  p.  223. 

I  can  see  no  good  reason  for  distinguishing  the  common 
Hare  of  the  Abyssinian  coast-land  from  the  desert  Hare 
of  Eg}^t.  It  is  a  small  greyish  animal,  very  slight 
in  the  body,  with  long  legs  and  very  large  ears.  It  has 
been   by  some   naturalists  looked  upon  as  the    Lepus 

T 


274  ZOOLOGY. 

habessiniciis  of  Hemprich  and  Ehi'enberg.  It  does  not, 
however,  agree  with  the  description  in  the  "  S3nnbol8e 
Physicee ; "  and  the  type  specimen  in  the  Berlin  Museum, 
which  I  have  examined  in  order  to  decide  the  question, 
is  certainly  a  diflferent  animal,  similar  in  size,  but  with 
small  ears  instead  of  unusually  large  ones,  these  being 
only  3f  in,  long  in  the  stuffed  skin. 

The  following  is  a  brief  description  of  the  type  spe- 
cimen of  L.  hahessinicus : — ^The  colour  grey,  mottled  with 
dusky,  as  usual,  especially  on  the  back ;  nape  pale  rufous, 
albescent  behind  the  ears ;  white  marks  at  side  of  face 
indistinct  General  size  and  build  similar  to  L.  CBgyptius, 
or,  perhaps,  a  little  larger.  It  approaches  still  more 
closely  to  the  specimen  of  L.  sinaiticus^  H.  and  E. 

The  foUowing  are  a  few  dimensions:— 

in. 

Length  of  face  from  between  ears  and   nose  measured 

over  carve  of  the  head 4 

Base  of  ear  in  frY>nt  to  nose,  in  a  direct  line 3^ 

Length  of  forearm  (radius)  and  carpus  frY>m  elbow-joint  to 

end  oftoe 5 

„        tibia 4j^ 

„        tarsus  from  tarsal  point  to  point  of  toe     ...  4^ 

„        ears,  as  above 3| 

The  following  are  dimensions  of  three  specimens  of 
Lepus  cBgyptius  shot  near  Zulla,  taken  from  the  animals 
before  skinning: — 

1.  2.  Male.      8.  Fenude. 

ft    in.       ft.    in.         ft.    in. 

Length  of  fSEtce,  frY>m  nose  to  between  ears    0    4  0    4        0    3j 
„      from   between  ears    to    top    of 

shoulder 0    3  0    3^      0    4^ 

„      from  top  of  shoulder  to  rump      .10  10        12 

„  of  tail,  to  end  of  hairs  .    ,    .    0    3j^  0    3^        — 

1  10^  1  10| 


MAMMALIA,  275 

1.  2.  Male.      3.  Female, 

(t.    in.       ft    in.         fT     In. 

Length  of  body  measured  in  a  direct  line 
along  the  side  from  front  of  shoulder 

to  rump 11        11        13J 

Height  at  shoulder —  0  10        0  11| 

Depth  of  body  behind  shoulder     ...  0    4]^ 

Girth  of  chest  behind  shoulder  ....      —  09        0    9| 

„     neck 0    6  —  0    7J 

Length  of  forearm  (radius),  from  elbow 

to  carpal  joint 0    4^ 

Length  of  carpus  to  point  of  toe  .    .    .  0    2^ 

of  middle  fore-toe 0    1^ 

of  thigh  (tibia) 0    6^ 

of  tarsus  to  point  of  toe    ...  0    4^ 

of  middle  hind-toe 0    1^ 

of  ear 0    6} 

Breadth  of  ear  in  the  middle  when  laid  flat  0    3^ 

If  this  species  be  not  Lepus  cBgyptius,  it  will  have  to 
be  named ;  but  I  can  see  no  difference  whatever,  after 
examining  several  specimens.  It  abounds  amongst  the 
bushes  on  the  shore  plain  about  Annesley  Bay,  and  is 
less  common  near  Massowa. 

Another  closely  allied  species,  with  paler  fur,  pro- 
bably a  variety  inhabiting  a  more  desert,  sandy  country, 
is  L.  isahellinus,  Elipp.,  which  is  identical  with  L,  (ethi- 
opicu^  of  Hemprich  and  Ehrenberg. 

ay.  L.  tigrensis,  W,  Blanf. 

Ann.  and  Mag.  Nat.  Hist.  1869,  p.  330. 
L.  ahyssinicusy  Lefebvre,  Atlas,  pi.  6,  fig.  1. 

L.  persimUiSf  L.  saxatili,  sed  minor,  cmida  brevioriy  plantarum  pilis 
ferrugineis,  hatid  umhrims. 

Fur  above,  grey,  mottled  with  black,  the  hairs  grey 
or  pale  isabelline  at  the  base,  then  for  a  short  distance 

T  2 


276  ZOOLOGY. 

black,  succeeded  by  isabelline  grey,  more  or  less  pale, 
the  tip  beiug  black  ;  towards  the  sides  the  black  portions 
gradually  die  out.  Flanks  and  breast  rufescent,  chin 
and  belly  white ;  face  isabelline,  mixed  with  black — 
whitish  around  the  eyes ;  ears  of  moderate  size,  dusky 
brown  externally,  a  fringe  of  isabelline  hairs  on  the 
outer  edge,  and  of  short,  dusky  hairs  at  the  tip,  inner 
edge  whitish ;  legs  isabelline,  mixed  with  black  exter- 
nally ;  soles  ferrugineous,  this  colour  extending  back  to 
the  tarsal  joint  on  the  hind-legs ;  tail  rather  long, 
white,  except  a  rather  broad,  dusky,  black  stripe 
above. 

Unfortunately  I  did  not  take  the  dimensions  of  this 
Hare  from  the  animal  before  skinning ;  the  following 
measurements  being  from  the  dried  skin,  the  length  of 
the  body  can  only  be  considered  approximate.  The 
lengths  of  the  head,  ears,  and  limbs  are  perfectly 
trustworthy,  however. 

1  s 

in.  in. 

Length  of  head  from  nose  to  base  of  ear  ....  4  3*8 

„        neck  and  body  from  ears  to  rump  (about)  14  13 

„        tail  3  in.,  hairs  at  end  1  in.      ....  4  4 

„        ears 6  6 

Breadth  of  ears 2*8  2*8 . 

Length  of  radius 4*4  4*5 

„        carpus  and  toes  to  end  of  claws    .     .     .  2'3  2*3 

„        tibia 5*3  6-2 

„        tarsus  and  toes  to  end  of  claws     .     .     .  4*5  4*5 

„        skull 3-65  3-45 

Breadth  of  skull  at  back  part  of  zygomata    ...  1*6  1'6 

Height  of  skull  from  base  of  lower  jaw  to  crown 

of  head 22  2-16* 

»  1  from  Takonda,  No.  287  ;  2  from  Adabagi,  693.    The  types  of  this,  as 
of  all  other  species  described,  are  placed  in  the  British  Museum. 


MAMMALIA,  -  277 

In  the  skull  of  Lepus  saxatilis  the  lower  jaw  is  higher 
in  proportion  to  the  length,  the  zygomatic  arch  much 
longer,  and  the  whole  skull  larger  and  more  powerful. 

Lepus  tigrensis  is  a  larger  and  heavier  animal  than 
L,  cegyptiuSy  L.  Iiabessinicus^  L.  sinaiticus,  &o.,  the  limbs 
being  nearly  twice  as  large,  and  the  whole  build  of 
the  animal  more  massive.  It  is  an  interesting  circum- 
stance to  find  that  the  common  Hare  of  the  Abyssinian 
highlands  approaches  more  neai'ly  to  a  South  Afidcan 
type  than  to  any  of  those  existing  in  the  surrounding 
countries,  and  this  aflSnity  is  in  accordance  with  the 
relations  of  the  avi-faima. 

It  is  not  at  all  surprising  that  both  this  Hare  and 
Lepus  cegyptius  from  the  coast  should  have  been  con- 
founded with  Hemprich  and  Ehrenberg's  L.  Imhessinicus, 
In  few  genera  are  the  specific  distinctions  more  difficult 
to  describe  than  amongst  the  Hares ;  and  I  should  not 
have  been  satisfied  that  Lepus  tigrensis  is  new,  had  I 
not  examined  Hemprich  and  Ehrenberg's  type.  I  still 
cannot  help  suspecting  that  there  is  some  error  in  the 
locality  assigned  to  the  type  in  question,  said  to  come 
from  near  Massowa,  or  that  its  remarkably  short  ears  are 
an  individual  peculiarity :  still,  if  it  be  a  monstrosity, 
it  certainly  belongs  to  the  Lepus  cegyptius^  or  a  closely 
allied  species,  and  not  to  the  race  inhabiting  the  Abys- 
sinian highlands.  It  is  rather  remarkable  that  the 
latter  should  have  been  so  long  overlooked. 

Lepus  tigrensis  abounds  around  Senafd,  Takonda, 
Adigrat,  Adabagi,  and  Antalo.  I  did  not  notice  it  in 
the   hill   country   of    Lasta.      Its   habits   are  precisely 


I  278  ZOOLOGY. 

those  of  the  European  Hare.  It  was  found  most 
commonly  on  open  ground  with  low  scattered  bushes, 
at  from  7,000  to  9,000  feet  above  the  sea,  I  believe 
that  the  few  hares  seen  in  the  Anseba  valley,  at  4,000 
to  5,000  feet  elevation,  belonged  to  this  species ;  but  I 
am  not  certain.  They  may  possibly  have  been  L.  habes- 
sinicus,  which,  in  that  case,  may  inhabit  the  sub-tropical 
region,  and  L.  tigrensis  be  confined  to  the  temperate. 

28.  Sciurus  annulatus,  Desm. 

Mammalogie,  p.  338. — Gray,  Ann.  and  Mag.  Nat.  Hist.,  Ser.  3,  yoL 

XX.  p.  329. 
Sc,  multicolor,  Riipp.  Neu.  Wirb.  p.  38,  t.  13.— Brehm,  Habescli. 

p.  62. 

1  have  considered  S.  multicolor  as  identical  with  S. 
annulatus  on  Dr.  Gray's  authority,  as  I  have  not  myself 
been  able  to  examine  a  sufficiently  full  series. 

I  only  met  with  this  Squirrel  in  the  thickets  on  the 
banks  of  the  Anseba,  where  it  was  not  very  rare  in  the 
lofty  trees.  It  is  a  quiet  little  animal,  less  lively  than 
most  squirrels. 

2g.  Xerus  rutilus,  Eiipp. 

Sciurus  rutilus,  Riipp.  Atlas,  t.  24. 
S.  xerus  hrachyotus,  Hemp,  and  Ehr.,  Symb.  Phys.,  t.  ix. 
Xerus  rutilans.  Gray,  Ann.  and  Mag.  Nat.  Hist.,  Ser.  3,  vol.  xx, 
p.  332, 

This  Ground-Squirrel,  which  is  figured  of  much  too 
rufous  a  colour  by  Ruppell,  is  not  rare  in  the  lower 
portions  of  the  passes  leading  to  the  highlands,  I  saw 
it  also  in  rocky  places  close  to  the  shore  of  Annesley 
Bay.     Frequently  five  or  six  are  seen  together  keeping 


MAMMJUA.  279 

to  the  ground  or  to  rocks^  and  taking  refuge  in  clefts 
and  holes,  never  upon  trees.  The  tail  is  very  long  and 
bushy. 

30.  X.  leuco-umbrinus,  Hupp. 

Biipp.  Neo.  Wirb.  p.  38,  in  remarks  on  the  genus  Sdurui. 

X  Mtosus,  Gray,  Ann.  and  Mag.  Nat.  Hist.,  Ser.  3,  yoL  xx.  p.  332. 

I  cannot  agree  with  Dr.  Gray  in  considering  the 
striped  Abyssinian  Ground-Squirrel  identical  with  the 
Cape  species.  RUppell  founded  his  species  on  the  dif- 
ference in  the  ears ;  the  Cape  Xerus  setosus  having  no 
external  conch,  whilst  the  Abyssinian  X.  Uuco-umhrinus 
has  one  distinctly  developed,  although  small  On  com- 
paring the  specimens  procured  by  myself  in  Abyssinia 
Avith  those  from  the  Cape  in  the  British  Museum,  I  find 
this  distinction  thoroughly  confirmed.  X.  setosuSy  more- 
over, appears  to  be  a  much  smaller  animal.  Specimens 
of  Xerus  from  Western  Africa  show  an  intermediate 
form  of  Ciu- ;  they  are  of  the  same  size  as  the  Abyssinian 
species,  but  much  darker  in  colour. 

Xerus  leuco-umhrinus  abounded  in  rocky  places  about 
Senafd  and  elsewhere,  in  Tigr(5  up  to  about  9,000  feet, 
and  1  found  it  as  low  as  4,500  in  the  Anseba  valley. 
It  has  six  mammae.  On  the  2d  of  March,  near  Takonda,  I 
shot  a  gravid  female  containing  four  well-grown  young ; 
two,  a  male  and  a  female,  in  each  horn  of  the  uterus. 

31.  Bathyergus  splendens,  Eiipp. 

Riippell,  Neu.  Wirb.  p.  36,  t.  12.— Brehm,  Habesch,  p.  63. 

This  peculiar  burrowing  rodent  is  very  closely  aUied 
in  form  and  habits  to  the  Bamboo  Rats  (Rhizomys)  of 


I 


280  ZOOLOGY. 

the  Himalaya  and  Indo-Chinese  countries.  It  burrowa 
in  meadows  or  grassy  glades  in  valleys,  throwing  up 
little  heaps  of  earth  like  molehills  at  the  entrances  to 
its  burrows.  It  appears  to  live  entirely  imder  the  earth ; 
and  although  I  several  times  saw  the  heaps  caused  by 
its  excavations,  I  only  obtained  specimens  of  the  animal 
on  the  borders  of  Lake  Ashangi,  where  they  abounded 
on  the  broad  grassy  plain  in  which  the  camp  stood. 

The  burrows  are  very  extensive,  and  the  galleries 
intricate  and  much  curved.  In  one  which  I  dug  outi, 
there  was  a  larger  chamber  in  the  centre  about  six 
inches  below  the  surface,  well  lined  with  dried  roots 
of  grass ;  from  this,  burrows  led  in  all  directions,  usually 
not  more  than  two  or  three  inches  below  the  surface  of 
the  ground,  but  occasionally  rather  deeper. 

I  twice  obtained  living  specimens,  one  of  which  I  kept 
for  some  days,  feeding  it  on  roots  of  grass,  which  it  ate 
freely,  and  which  I  have  no  doubt  are  its  natural  food. 
In  this  also  there  is  a  close  resemblance  to  Rhizomys^ 
which  lives  on  roots  of  bamboos. 

Bathyergus  splendens  is  a  sluggish  animal,  with 
evidently  imperfect  sight,  and,  despite  its  formidable 
teeth,  it  was  not  diflBcult  to  capture,  when  its  under- 
ground retreat  was  cut  off.  The  fur  is  beautifully  fine 
and  soft. 

The  following  dimensions  arc  from  a  specimen  in 
spirit : — 

In. 
Length  of  head  and  body 10 


» 


of  tail 3 


„      from  :ear  to  nose I'T 


MAMMALIA.  2fi\ 

in. 

Length  of  fore  leg 2 

„  sole  of  fore  foot 1 

„  middle  toe  and  claw  of  foi^  foot 0*5 

Breadth  of  sole  of  fore  foot 0*45 

Length  of  hind  leg 2*5 

„  sole  of  hind  foot 1*26 

),  longest  toe  (2d)  of  hind  foot 0*45 


32.  Pectinator  Spekei,  Blyth. 

Blyth,  Joor.  As.  Soc.  Bengal,  1855,  yol  xxiy.  p.  294. 
HeugL  Noya  Acta  Leopoldino-Car.  Acad.  1861,  p.  1,  t.  2. 

This  very  remarkable  rodent  was  first  discovered  by 
the  late  Captain  Speke  in  the  Somali  country,  and  is 
one  of  the  most  interesting  animals  which  I  obtained. 
I  first  procured  it  on  some  hills  near  Hadoda,  barely 
1,000  feet  above  the  sea,  and  I  found  it  common  in  the 
pass  leading  to  Senaf^,  at  elevations  below  about  4,000 
or  5,000  feet. 

It  inhabits  rocky  clifis  precisely  as  Ilyrax  does,  hiding 
when  disturbed  in  crevices.  It  is  social,  several  indi- 
viduals being  usually  seen  about  one  spot,  lying  upon 
ledges  of  rock  or  projecting  stones.  Unless  approached 
cautiously,  all  hide  themselves  amongst  the  stones, 
emerging  again  after  a  time.  The  spots  haunted  by 
these  little  animals  may  usually  be  recognised  by  the 
quantity  of  dried  pellets  of  dung  which  lie  about. 
These  are  elongated,  and  rather  larger  than  those  of 
rats. 

The  appearance  of  Pectinator  Spekei  is  much  that  of 
a  small  dark  grey  squirrel  with  a  singularly  short  tail, 
which  it  has  a  very  squirrel-like  trick  of  jerking  over  its 


282  ZOOLOGY. 

back  It  is  usually  seen  lying  flat,  and  is  with  difficulty 
distinguished  from  the  rock  around.  The  fur  is  very 
woolly  and  soft,  and  the  skin  remarkably  thin  and 
tender ;  so  much  so,  that  I  never  remember  experiencing 
the  same  amount  of  difficulty  in  skinning  any  other 
mammal  or  bird.  The  skin  tears  in  the  fingers  like  wet 
tissue  paper. 

In  the  stomach  I  invariably  found  vegetables  alone, 
much  chewed  and  broken  up,  so  that  it  was  impossible 
to  say  what  kinds  it  eats ;  but  evidently  its  food  consists 
of  leaves.  I  suspect  that  it  feeds  at  night ;  I  never  saw 
it  doing  so  in  the  daytime,  and  far  more  anjmals  arc 
seen  out  in  the  evening  than  during  the  day. 

The  following  are  the  measurements  of  a  male  shot  at 
Sum  (1),  and  a  female  killed  near  Undul  Wells  (2), 
taken  on  the  animal  immediately  after  death  : — 


1 

In. 

2 
In, 

2 

— 

2-6 

5-5 

6*75 

2-25 

2 

1-35 

1*5 

Length  from  muzzle  to  between  ears 
,,      from  muzzle  to  nape      .    . 
„      from  between  ears  to  rump 
„      from  nape  to  rump   .    .    . 

„      of  tail 

„      of  bairs  beyond  end  of  tail 

Total  length  from  nose  to  end  of  tail  hairs, 

measured  along  curve  of  the  back    .    .    .11-10        1175 

Length  of  fore  foot  from  heel  to  end  of  claws    .  0*8  0*75 

„        middle  toe  and  claw  of  fore  foot .    .  0*38  0*45 

„        hind  foot  from  heel  to  end  of  claws  1*4  1*5 

„        middle  toe  and  claw  of  hind  foot    .  0*5  0*5 

Longest  whisker 3  3*5 

Height  of  ear 0*5  ()-5 

Breadth  of  eai 0*5  0-5 

Diameter  of  eye-baU,  about  \  inch. 


MAMMAUA.  283 

The  general  colour  is  greyish  brown,  with  whitish 
marks  over  the  eyes,  and  before  and  behind  the  base  of 
the  ears.   Some  of  the  lower  moustachial  hairs  are  white. 

Some  of  the  specimens  of  this  animal  in  spirits  pro- 
cured by  Mr.  Jesse,  have  been  sent  to  Dr.  Peters,  who 
is  preparing  a  complete  description  of  them  for  the  Zoo- 
logical Society  of  London. 

The  figure  in  the  "  Nova  Acta  "  gives  a  very  poor  idea 
of  this  animal,  far  worse  than  Blyth's  original  description. 
It  is  much  too  rufous  and  pale.  Indeed,  it  differs  so 
much,  that  I  thought  I  had  a  distinct  species,  untU  I 
compared  my  specimens  with  the  original  type  in  the 
Indian  Museum  in  Calcutta. 

33.  Mus  abyssinicus,  Eiipp. 

Riipp.  Mus.  Senck.  ill.  p.  104,  t.  viL  f.  1. 

This  is  not  a  true  MuSy  but  appai-ently  no  genus  has 
been  proposed  for  the  little  group  to  which  it  belongs. 
A  very  similar  animal  from  Mosambique  is  the  t)^e  of 
the  genus  Pelomys  of  Peters ;  but  the  dentition,  as 
pointed  out  by  that  naturalist,  shows  important  dif- 
ferences. 

Mas  cd)i/ssuticu6  wat)  found  by  Ruppell  high  on  the 
mountains  of  Samyen.  I  met  with  it  abundantly  above 
10,000  feet  on  the  Wandaj  pass,  and  the  Wadela  plateau, 
near  Santara.  On  the  lower  parts  of  the  plateau,  near 
Esindy4  I  did  not  observe  it.  It  inhabits  grassy  plains, 
which  are  burrowed  in  all  directions  by  its  holes,  and  the 
little  animals  may  be  seen  by  dozens  running  about.  It 
appears  peculiar  to  the  subalpine  fauna. 


284  zooLOor. 

34.  %  Mus  dembeensis,  Riipp. 

Riippell,  Mas.  Senck.  iiL  p.  109,  t.  yL  f.  3. 

T  obtained  two  specimens  of  a  mouse  at  Ashangi, 
which  only  differ  from  RuppcU's  description  and  figure 
in  the  tail  being  shorter  than  the  body  instead  of  longer. 
It  may  perhaps  be  distinct. 

35.  Mus,  sp.  1 

Two  specimens,  Adigrat. 

36.  Mus,  sp. 

I  have  been  quite  unable  to  identify  these  two  species, 
but  I  have  not  compared  them  sufficiently  to  venture  to 
describe  them  as  new. 

37*  Dipus  gerbillus,  Oliv.  (nee  Meriones  gtrUUtu^  Eiipp.). 
OHt.  BoU.  de  la  Soc.  Philom. — Desmarest,  Mamm.  p.  321. 

Very  common  close  to  the  coast  at  ZuUa.  Meriones 
gerhillus  of  Ruppell  appears  to  be  a  much  larger  species. 


AFES.  285 


Class  AVES. 

Ordee  RAPTORES. 
Family    VULTURID^ 

X.  Gyps  Rueppelli,  Bp. 

Vultur  KoUni,  Biipp.  Atlas,  fig.  32,  nee  Daadin. 

F.  BuppeUif  Bp.  Rev.  and   Mag.  Zool.  1850,  p.   477.— Brehm, 

Naumannia,  1862,  Pt.  XL  p.  44. 
Oyps  fvXvus,  avis  adulia,  Riipp.  Syst.  Uebers.,  No.  4. 
O,  magnifietiSj  V.  Miill.,  Descr.  Nouv.  Ois.  d'Afr.,  Tab.  V. 
O,  Riippdli,  V.  Heugl.,  Omith.  N.  0.  Africa's,  No.  2,  p.  6. 

The  older  birds  of  this  species  may  always  be  dis- 
tinguished by  the  broad  silver-grey  edgings  to  the 
back  feathers,  and  especially  to  the  wing-coverts.  So 
marked  is  this  that  some  birds  have  a  very  handsome 
mottled  appearance,  the  grey  even  predominating.  I 
unfortimately  could  not  collect  many  specimens,  and  not 
being  aware  at  the  time  of  the  interest  attaching  to  this 
particular  race  I  did  not  especially  notice  its  distribution 
relatively  to  Gyps  fulvus.  I  am,  however,  of  opinion 
that  the  great  majority  at  least  of  the  birds  which  are 
met  with  on  the  Abyssinian  highlands  belong  to  the 
present  species.     I  find  a  note  in  my  field-book  men- 


286  ZOOLOOr. 

tioning  the  great  prevalence  of  mottled  individuals  on 
the  high  plateaux  of  Wadela  and  Dalanta.  At  the 
same  time.  Gyps  Rueppelli  is  by  no  means  confined  to 
the  higher  portions  of  the  table-land.  .  I  met  with  it 
abundantly  in  the  Anseba  valley,  at  4,000  to  4,500 
feet  above  the  sea,  and  a  large  number  congregated 
around  the  carcase  of  a  cow  killed  by  a  lion  at  Rairo, 
north  of  the  Lebka  valley,  at  an  elevation  of  only 
3,000  feet.  But  from  the  circumstance  that  I  did  not 
especially  notice  the  conspicuously  mottled  feathers,  I 
am  inclined  to  suspect  that  the  majority  at  least  of 
the  enormous  number  of  Griffon  Vultures  which  were 
attracted  to  the  camp  at  Komayli,  at  the  base  of  the 
hills,  belonged  to  G.fulmis.  At  Zoulla,  on  the  shores 
of  Annesley  Bay,  despite  the  large  number  of  carcases. 
Griffon  Vultures  were  not  nearly  so  common  as  a  few 
miles  further  inland. 

The  figure  in  Ruppell's  *'  Atlas ''  gives  by  no  means  a 
good  idea  of  this  bird ;  the  bill  is  not  yellow,  nor  the 
neck  flesh-coloured :  the  former  is  dusky,  the  latter  dull 
ashy.  Heuglin  (Omith.  N.  0.  Afr.)  has  described  them 
correctly.  The  plumage,  too,  is  coloured  far  too  brown 
by  Ruppell. 

I  frequently  saw  Otogyps  auricularis^  Daud.,  but  did 
not  preserve  a  specimen.  It  is,  like  its  Indian  represen- 
tative, 0.  calvuSy  Scop.,  much  rarer  than  other  Vultures, 
two  or  three  occurring  amongst  a  large  number  of  the 
common  species.  I  met  with  it  on  the  highlands  at  a 
considerable  elevation,  and  I  occasionally  saw  it  on  the 
Anseba  and  at  lower  levels. 


AFE8.  287 

2.  Neophron  percnopterus  (linn.). 

Rachamah  of  Brace,  Trayels,  toL  v.  p.  163,  with  plate. 
Riipp.  Syst.  Uebers.  No.  2. — Lefebvre,  p.  46. — Brehm,  Habesdi, 
2.— Heugl.  Omith.  N.  0.  Africa's,  No.  7. 

Habits  and  appearance  precisely  like  those  of  the 
Indian  variety.  Common  everywhere  from  the  sea- 
level  up  to  10,000  feet,  and  equally  abundant  near 
the  camps  on  the  Wadela  plateau  and  on  the  shores  of 
Annesley  Bay. 

3.  N.  pileatus  (BurcheU). 

VuUur  pUeatuSf  Burch.  Trav.  S.  Afir.  ii.  196.^ 

Cathartes  monachuSf  Temm.  PI.  coL  222. 

Ne<yphron  pUecUus,  Biipp.  Syst.  Uebers.  No.  3. — Lefebvre,  p.  46. — 

Brehm,  Habesch,  No.  1. — HeugL  Omith.  N.  0.  Afr.  No.  8. 
N.  mwuuhus,  Ferret  et  Gall.  No.  2,  p.  178. 

This  is  a  bird  of  very  different  habits  from  N.  per- 
cnopterus,  and  far  more  vulturine  both  in  its  flight  and 
food.  Numbers  usually  collect  around  a  carcase,  which 
is  very  rarely,  if  ever,  the  case  with  N.  percnopterus. 
The  difference  is  best  shown  in  the  fact  that  both  Euro- 
peans and  Eastern  people  frequently  speak  of  N.  percnop- 
tcrus  as  a  Kite,  whilst  no  one  could  ever  consider 
N.  pileatus  as  anything  else  than  a  Vulture.  I  greatly 
doubt  the  propriety  of  ranking  both  in  one  genus. 

N.  pileatus  was  the  only  Vulture  common  about  the 
camp  at  Malkatto,  on  the  shores  of  Annesley  Bay.  It 
abounded  on  the  Anseba,  and  occurred,  but  less  abun- 
dantly, on  the  highlands. 

^  As  all  naturalists  identify  this  species  with  Burchell's  description,  I 
suppose  there  can  be  no  question  of  its  identity ;  yet  BurcheU  expressly 
states  that  his  species  was  equal  in  size  to  the  laigest  African  yultores, 
and,  in  the  description,  adds,  "  Species  inter  majores." 


288  ZOOLOGY. 


Family  FALCONIDiR 
4.  Palco  barbams,  L. 

SalTin,  Ibis,  1869,  p.  184,  pL  «.— Heog.  Orn.  N.  O.  Afr.  Ka  IL 
F.  peregrinoidu,  B,upp.  Sjst  Uebersw  Na  25. 

I  shot  a  single  female  specimen  of  a  Falcon  in  the 
Anseba  valley,  which  appeared  to  me  too  large  for  this 
species,  and  Mr.  J.  H.  Gumey,  to  whom  I  submitted  it, 
agreed  with  me  at  first  in  considering  it  a  specimen  of 
F.  minor  J  Bon.     It  is  a  comparatively  young  bird,  just 
changing  the  brown  immature  plumage  for  the  greyer 
livery   of    the  adult     The  wing  measures  nearly   13 
inches,^  tail  6 '5,  tarsus  2,  middle  toe  without  claw  2. 
The  length  of  the  wing  in  Falco   barbarus   is  given 
by  Mr.  Salvin  ("Ibis,''  voL  i.  p.  189)  as   11  to  11-25 
inches   (specimens   supposed   to    be  females) ;   by  Von 
Heuglin  ("Orn.  N.  0,  Afirica's,"  p.  22)  as  10^' 4''^  to 
10''  6'"  (French  measure)  in  the  male,  and  11"  6'"  to 
11^'  9'"  in  the  female  (about  11  and  12  J  inches  English 
respectively) ;   whilst  that  of   F.  minor  is   given   by 
Layard  ("Birds  of  South  Afidca,"  p.  19)  as  12^  in  the 
male,  13   in   the   female.     The  tail,  however,  is  much 
longer  in  F.  minor,  8  to  8f  inches  according  to  Layard ; 
and  Mr.  Gumey,  on  comparing  the  Abyssinian  specimen 
with   a  fully  adult    female  F.  barbarus,   fix)m  North 
Africa,  finds  that  there  is  no  important  difierence  in  size. 

The  chin  and  the  upper  part  of  the  breast  are  white, 
the  lower  breast  and  abdomen   stroDgly  rufous,  with 

^  In  both  wings  the  longest  feather  is  slightly  imperfect  at  the  end,  the 
absolute  length  being  12  J  inches. 


JFES.  289 

rather  close  imperfect  black  transverse  bars ;  cbin-stripe 
very  much  developed,  and  a  tendency  to  a  rufous  collar 
at  the  back  of  the  neck. 

5.  P.  tanypterus,  lichi 

Tinnuncvlu$  biarmiett$,  Schleg.  Abhand.  t  12,  13. — Riipp.  Syst 

Uebere.  No.  30. 
Falco  hiarmicM,  Ferret  et  OaUinier,  No.  4. — Lefebvre,  p.  67, 
F,  cerviccUiSf  HengL   Syst.    Uebers.   No.    44. — Brehm,  Habesch, 

No.  12. 
F,  lanarius  nubicuSy  Heugl.  Orn.  N.  0.  Afr.  No.  12,  p.  23. 

The  common  Falcon,  I  believe,  of  the  Abyssinian  high- 
lands. I  shot  three  specimens  on  the  plateau;  one  at 
10,500  feet,  on  the  top  of  the  Wandash  saddle,  another 
at  Dongolo,  a  third  at  Adigrat.  This  bird  was  also 
frequently  met  with  in  the  Anseba  and  Lebka  valleys. 

in.  in. 

In  the  male  firom  Wandash  the  wing  measures  .  13*6  ;  tail,  7*5. 
In  a  female  from  the  Lebka  river         ,,  .  15        „     8. 

The  latter  specimen  is  very  rufous  beneath,  with  very 
few  black  marks. 

6.  F.  sacer,  auct.   (?  Gm.)  ^ 

Gould's  Birds  of  Asia,  pt.  xx. — Schlegel,  Traits  de  Fauconnerie, 

pi.  5. 
F,  saqer,  Heugl.  Cm.  N.  0.  Afr.  No.  13. 

Iris  brown;  orbit,  cere,  and  nearly  the  whole  beak 
bluish  grey,  tip  of  bill  dusky ;  legs  pale  greenish  grey. 

I  killed  a  single  specimen  at  Kelamet  in  the  Lebka 
valley,  nearly  3,0.00  feet  above  the  sea.     I  saw  a  pair 

^  I  greatly  doubt  if  this  be  the  Falco  tacm-  of  Gmelin. 

U 


290  ZOOLOGY. 

chased  by  crows  {Corvus  scapulatus)^  and  succeeded  in 
shooting  one.  It  has  a  remarkably  pale  head,  whitish 
isabelline  with  dusky  streaks,  is  dusky  brown  above, 
and  dusky  with  broad  pale  edges  to  the  feathers  beneath, 
darker  on  the  body  and  lighter  on  the  head  than  in 
Schlegel's  figure.  The  wing  measures  14  inches,  tail 
nearly  8. 

I  saw  a  very  similar  specimen  at  Lahej,  near  Aden, 
which  was  shot  by  a  friend.  Persian  and  Afghan 
specimens,  Dr.  Jerdon  informs  me,  have  darker  heads. 

7.  P.  {Tvnnuncului)  tinnunculus,  Linn. 

Ferret  et  GalL  No.  6.— HengL  Om.  N.  0.  Afr.  No.  21. 
TinnuncvXus  alauda/rius,  Biipp.  Syst.  Uebers.  No.  27. 
Lefebyre,  p.  68. 

Common  both  in  the  highlands  and  lowlands  in  the 
winter  and  spring.  The  bird  abounded  on  the  former  as 
late  as  April  None  were  observed  in  the  Anseba  valley 
in  July  and  August. 

A  very  dark-coloured  female  from  Adigrat  shot  on  the 
23d  April  is,  Mr.  Gumey  informs  me,  similar  to  the  race 
inhabiting  Madeira.  Von  Heuglin  states  that  the  birds 
which  breed  in  Northern  Africa  are  more  deeply  coloured 
than  those  which  migrate. 

8.  F.  (TitinunctUus)  cenchris,  Naum. 

HengL  Oni.  N.  0.  Afr.  No.  23. 

TinnunculuB  emchrU,  Biipp.  Syst.  Uebers.  No.  28. 

Not  rare  on  the  Abyssinian  highlands. 


AVES.  29 1 

9.  Melierax  polyzonus,  Biipp. 

FaUo  (Nisus)  polyxontu^  Biipp.  Neo.  Wiib.  p.  36,  tftf.  15,  fig.  1. 

Melierax  polyzonusy  Biipp.  Syst.  Ueben.  No.  43. 

Niitu  polyzontu,  Fer.  et  GalL  No.  7. 

Melierax  mueicuSf  Lefebyre,  p.  70. 

Attur  polyzonuSf  HeagL  Om.  N.  0.  Afr.  No.  39. 

It  is  darloBh  brown,  orbit  dull  yellow,  cere  and  legs 
bright  orange,  claws  dusky,  base  of  bill  orange,  passing 
into  flesh  colour,  tip  dusky. 

Common  on  the  searcoast  and  in  Habab  up  to  about 
3,000  feet  elevation,  but  not  seen  in  the  Anseba  vaUey, 
or  at  any  high  elevation,  and  appears  chiefly  restricted 
to  the  tropical  zone.  Heuglin,  however,  states  that  he 
has  met  with  it  at  greater  heighta.  It  is  usually  seen 
perched  on  a  tree,  less  frequentiy  on  the  ground,  its 
food  being  reptiles  and  insecta  Its  flight  is  very 
buzzard-like  and  not  swift,  and  it  generally  settles 
.gain  .iter  i.yi,g  .  short  di^c. 

ID.  Nisus  tachiro,  Daud. 

Faleo  taehiroy  Baud.  Traits  Comp.  d'Omith.  ii.  90. 
F,  (Astur)  undv>liventery  Biipp.  Neu.  Wirb.  p.  40,  t  18,  fig.  1. 
Nistu  unduliventer,  Biipp.    Syst.   Uebers.  No.  42.^HeugL  Om. 
N.  0.  Afr.  No.  43. 

Iris  yellow ;  skin  round  eyes,  cere,  gape,  and  legs  also 
bright  yellow.  Both  M.  Jules  Verreaux  and  Mr.  Gumey, 
two  of  the  best  authorities  on  South  African  Raptores, 
consider  the  only  specimen  which  I  obtained  of  this 
bird  as  identical  with  N.  tachiro.  My  single  skin  is 
that  of  a  very  old  male  in  superb  plumage.  All  the 
upper   parts   are  dusky  with  a  slaty  tinge,  tail  dusky 

1X2 


292  ZOOLOGY. 

black,  all  the  feathers  except  the  outermost  with  four 
white  bands,  the  anterior  of  which  is  concealed  by  the 
coverts,  and  a  narrow  white  tip,  quills  obsoletely  banded 
with  dusky  above,  distinctly  with  white,  especially  on 
the  inner  webs,  below.  All  the  under  parts  closely 
banded  with  ferruginous,  except  the  throat,  where  the 
stripes  are  grey,  the  under  tail-coverts,  which  are  pure 
white  with  only  traces  of  bands,  and  the  thigh-coverts, 
which  are  pure  ferruginous.  I  can,  however,  detect 
slight  traces  of  banding  even  on  these.  The  measure- 
ments are,— whole  length  about  15  inches,  wing  7*8, 
tail  6 '5,  tarsus  2*4.  There  is  a  specimen  of  N.  tachiro 
from  South  Africa  in  the  British  Museum,  which  agrees 
well  with  that  from  Abyssinia  in  colouring. 

I  only  once  met  with  this  bird,  which  must  be  very  rare 
in  Abyssinia,  as  both  Ruppell  and  Von  Heuglin  seem  also 
to  have  obtained  it  only  on  one  occasion  each,  the  latter 
procuring  three  specimens.  I  found  a  pair  at  Goona 
Goona,  near  Senafd,  in  the  middle  of  March,  evidently 
breeding,  chasing  each  other  through  bushes  beneath  a 
small  waterfall,  but  I  only  succeeded  in  shooting  the 
male. 

XX.  Nisus  (Micronisus)  niloticus,  Sund. 

Mdierax  gabar^  Riipp.  Syst.  Uebers.  No.  44. 

M,  nilotictu,  Sundevall,  Ofvers.  E.  Vet.  Akad.  Forh.  1850,  p.  132. 

Nisut  gabar,  HeugL  Orn.  N.  0.  Afir.  No.  46. 

Iris  dull  red ;  cere  and  legs  orange  :  bill  dusky. 

I  shot  a  pair  of  these  birds  at  Ailat,  west  of  Massowa. 
They  appear  to  me  to  diflFer  conspicuously  from  the 
southern  species  N.  gabar,  Daud  in  the  less  amount  of 


AFES.  293 


grey  on  the  breast,  and  the  much  fewer  transverse  bands 
on  the  abdomen.  The  tail  also  appears  slightly  shorter. 
The  foUowing  are  the  measurements  of  the  two  speci- 
mens, both  apparently  females : — 

to.  in. 

1.  Wing,  7-76 ;  taU,  6*6. 

2.  „      7-8       „    6-8. 

The  somewhat  shorter  tail  is  the  character  upon 
which  SundevaU  grounds  his  distinction  of  the  species.^ 
Mr.  Gumey,  who  kindly  examined  one  of  my  specimens, 
considers  it  distinct.  Von  Heuglin  also  appears  inclined 
to  separate  the  northern  race. 

12.  N.  {MicroniruB)  niger,  VieilL 

Bj^arviiis  niger,  YieilL  Enc.  p.  1269. 

Circus  ma/urusy  Biipp.  Syst.  Uebers.  No.  47  ;  teste  HengL  sed  ? 

Nisus  niger,  HeugL  Om.  N.  0.  Afr.  No.  47. 

Iris  brown  ;  cere  and  bill  black ;  legs  yellow  with  black 
patches  in  front,  and  also  above  the  toes ;  nails  black. 

I  shot  two  males  together  in  jungle  near  the  source  of 
the  Lebka.  Both  had  been  feeding  on  either  birds  or 
small  mammals,  not  on  insects. 

By  some  authors  this  is  considered  a  mere  melanism 
of  N.  gahar.  It  appears  to  me  to  differ  from  the 
northern  race,  N.  niloticus,  in  smaller  size  and  in  the 
markings  on  the  underside  of  the  quills  being  larger 
and  less  numerous.  The  quills  themselves  also  appear 
rather  different   in  shape.      The  dimensions   are,  wing 

^  Sundevall  says  certainly  "  M,  gabar  ex  Senncutria  a  Caffro  differt  eauda 
multo  hngiore  fere  gradaJta^  but  his  measurements  show  that  it  is  the 
northern  or  Sennaar  form  which  has  the  shorter  taiL 


294  ZOOLOGY. 

675  in.,  tail  5*5,  but  the  tail-feathers  appear  to  be  not 
quite  fully  grown. 

13.  N.  lMicronisu8)  sphenurus,  Riipp. 

FaUo  (Nistui)  gphenuniSf  Riipp.  Neu.  Wirb.  p.  42. 
Aecipiter  braehydactyluSf  Swains.  B.  West  Afr.  voL  L  p.  118. 
Nisus  gphenurus,  Riipp.  Syst.  Uebers.  No.  41,  p.  6,  t.  2. 
MicronisuB  guUatus,  Heugl.  Joum.  f.  Omith.  1861,  p.  430. 
Nisus  hadiuSf  HeugL  Om.  N.  0.  Afr.  No.  45. 

Iris  yellow,  or  greenish  yellow  in  young,  orange  or 
scarlet  in  old  birds ;  cere  yellow  ;  legs  dark  yellow ;  bill 
black.  In  young  birds  the  breast  is  spotted,  in  adults 
banded,  the  changes  precisely  as  in  N.  hadius  (C£ 
Jerdon,  B.  Ind.  vol.  L  p.  48). 

When  I  saw  this  bird  in  Abyssinia,  it  always  appeared 
to  me  very  much  smaller  than  the  common  Indian 
Shikra  {N.  hadius^  Gm.),  and  the  measurements  appear 
to  bear  out  this  view,  although  Schlegel  and  Von 
Heuglin  unite  the  two  races.  The  following  are  measure- 
ments of  three  specimens  of  N.  sphenurus : — 

LeiigtlL     Wing.       Tail.       Tumis. 
In.  in.  in.  in. 

Old  female 12         7*4        66        17 

Young  female 1275    7*6        6*2        1*7 

Young  male 11         7  5-3        1*6 

These  agree  very  fairly  with  RUppell's  and  Von 
Heuglin's  measurements.  But  the  measurements  of  the 
Indian  N.  hadius^  according  to  Jerdon,  are  : — 

Ltngth.        Wing.      lUL  Tuvna. 

in.  in.  in.  in. 

Female 14  to  15        8^        7        nearly  2 

Male 12i  7^        5^ 

N.  sphenurus  was  extremely  abundant  amongst  Tama- 
risk trees  at  Mohabar  and  elsewhere  in  the  Lebka  valley. 


AFES.  295 

and  far  from  rare  in  the  Anseba.  BirdB  in  the  young 
plumage  were  more  numerous  than  adults.  All  at  Moha- 
bar  were  living  on  CicadcB,  and  so  fiEu:  as  my  observations 
extended,  this  bird  is  very  much  more  insectivorous 
than  its  Indian  representative. 

Mr.  Gumey  informs  me  that  the  West  African  N.  bror 
chydactylus  of  Swainson  is  identical  with  N.  sphenurus. 

14.  Aquila  rapax,  Temm. 

Faleo  rapax,  Temm.  PL  ooL  422  (a  veiy  rufous  specimen). 

Aquila  aUncans,  Riipp.  Neu.  Wirb.  p.  34,  t  13,  figs.  1  and  2. — 
Ferret  et  Gallinier,  No.  9. — ^Des  Murs,  p.  65. 

A.  rapaxy  Riipp.  Sjst.  Uebers.  No.  15. — Heugl.  Syst  Uebers.  Nos 
16, 18, 19,  20  (partly  under  MS.  names). — Brehm,  Habesch, 
Nob.  7  and  8.~Heu^  Om.  N.  0.  Air.  No.  45. 

A,  navioideSy  Cut. 

Iris  yellowish  brown,  or  yellowish  grey  much  mottled 
with  brown  ;  gape,  cere,  and  legs  yellow  ;  claws  black ; 
bill  dusky. 

A  somewhat  larger  and  much  more  powerful  bird,  with 
conspicuously  stronger  beak  and  legs,  than  the  closely 
allied  Indian  A.fulvescens  of  Gray.  The  plumage  varies 
from  umber  brown  to  rufous,  the  latter  colour  prevailing 
in  adult  birds,  especially  on  the  head  and  upper  part  of 
the  back.     Old  birds  are  whitish  (A.  albicans,  Riipp.). 

This  bird  is  common  everj^where  in  Abyssinia ;  I  ob- 
tained specimens  both  on  the  highlands  and  in  the  low 
plains  of  Samhar  close  to  the  coast.  I  shot  one  sitting 
on  a  small  rocky  hill  in  the  latter  in  August.  It  breeds 
on  the  highlands  about  January.  I  saw  one  nest  on  the 
top  of  an  isolated  tree  near  a  village,  containing  a  young 
bird  almost  full-grown,  on  March  24tL     Several  of  these 


296  ZOOLOGY, 


Eagles  frequently  collected  about  carcases  in  company 
with  Vultures,  Neophrons,  and  Crows. 
I  add  measurements  of  two  specimens : — 


Wlog. 
in. 

Tail 
in. 

Bill  to  Oape. 
in. 

Tamit. 
in. 

Middle  toe 
withuat  claw, 
in. 

1.  Male  (Senaf(6,  Tigr^)    . 

20-3 

11 

2-5 

3*4 

2-3 

2.  Female  ?  ( Amba,  near 

Massowa,  Samhar) 

21 

11 

2-4 

3-5 

2-4 

15.  Helotarsus  ecaudatus,  DaucL 

FaXco  eeattdatuSf  Baud.  Traits  Comp.  d'Omith.  ii.  64. 

HdotarstLS  ecaudatus  et  leuconotuSf  Riipp.  Syst.  Uebers.  23, 23a. — 
Lefebvre,  p.  66.— Atlas,  pi.  2  ( jun.). — HeugL  Syst.  Uebers.  35 
and  36.— Brehm,  Habesch,  No.  10.— HeugL  Om.  N.  0.  Afr. 
No.  60. 

Iris  brown ;  cere  (large)  and  lores  Indian  red,  with 
some  yellow  around  and  above  the  nostrils ;  bill  deep 
yellow;  tip  dusky;  legs  mottled  pale  Indian  red  and 
yellow,  claws  dusky. 

By  no  means  a  common  bird  in  Abyssinia :  I  only 
killed  one  which  was  flying  over  me,  in  the  Anseba 
valley.  Mr.  Jesse  shot  a  second  on  the  ground  near  our 
camp.  Another  specimen,  which  was  shot  at  Mayen, 
was  given  to  me  by  Captain  Sturt  of  the  Transport 
Train.  I  occasionally  saw  birds  of  this  species,  especially 
about  Senafd,  soaring  at  a  great  height,  their  pointed 
wings  and  extremely  short  taU  rendering  them  no  less 
conspicuous  than  the  contrast  between  the  white  under- 
side of  the  wings  and  the  black  body.  The  flight  is 
superb,  more  like  that  of  a  vulture  than  of  an  eagle, 
as  the  head  sweeps  along  with  motionless  wings,  oc- 
casionally high  up  in  the  air,  but  more  frequently  at 


AVES.  297 

about  150  to  200  feet  above  the  ground.     I  never  saw 
it  near  the  coast. 

x6.  Buteo  augur,  Hupp. 

Abyssinian  white-breasted  Lanner.    Salt,  App.  p.  xliL 

Faieo  (BvJteo)  augur  €t  hydrophilus,  Riipp.    Nea.   Wirb.  p.  38, 

t  16, 17. 
Buteo  augur,  Riipp.  Syst.  Uebers.  No.  10. — Ferret  et  GalL  No.  11. 

— Brehm,  Habesch,  No.  6. — HeugL  Cm.  N.  0.  Afr.  No.  69, 

Iris  yellowish  brown  ;  cere  and  legs  yellow. 

This  very  handsome  Buzzard  appears  to  belong  to  the 
temperate  and  subalpine  fauna  of  Abyssinia.  It  is  very 
rarely  met  with  below  6,000  to  7,000  feet,  and  I  never 
saw  it  on  the  Anseba  or  in  the  passes  below  Senafd.  At 
the  higher  elevations  its  numbers  increase  greatly,  and  it 
abounds  on  the  Wadela  plateau  at  10,500  feet. 

It  is  more  often  seen  perched  on  the  ground  than  on 
trees,  and  rarely  flies  to  any  distance  when  disturbed. 
A  female  shot  at  the  end  of  February  had  the  ovaries 
greatly  enlarged  and  was  evidently  breeding,  but  I  did 
not  see  the  nest.     It  is  by  no  means  a  wary  bird. 

I  am  rather  of  opinion  with  Ruppell  that  the  dark- 
coloured  birds  are  young  and  not  a  melanoid  variety. 
I  shot  two  black  specimens ;  one  evidently  immature,  the 
other  apparently  a  l)ird  of  the  year. 

17.  Gypogeranus  serpentarius,  Gm. 

Faico  serpentarius,  Gm.  Syst.  Nat.  L  250. 

Serpentarius  reptUivoruSy  Riipp.  Syst.  Uebers.  No.  49. — Des  Murs 

p.  70. 
S.  typieuSy  Ferret  et  Gall.  No.  16. 
(hfpogeranus  serpentarius,  Heugl.  Om.  N.  0.  Afr.  No.  49. 

Iris  grey ;  legs  yellowish. 


298  ZOOLOGY, 

A  raxe  bird  on  the  Abyssinian  highlands :  I  only  saw 
it  twice,  and  obtained  one  specimen.  In  the  stomach 
were  locusts  and  lizards ;  the  latter,  of  which  there  were 
at  least  a  dozen,  were  all  scinques  of  two  species,  Gongyltis 
oceUatus  and  Uuprepeis  qumquetceiiiatus. 

The  pace  at  which  the  Secretary  bird  can  walk  is 
astonishing.  The  bird  above  mentioned,  after  being 
wounded,  walked  away  as  fast  as  I  could  run  after  it 

z8.  Gypaetos  meridionalis,  Keys  and  Bias. 

NiBser  or  Golden  Eagle,  Bruce,  yoL  y.  p.  155 ;  Niaser  Werk  on 

plate. 
African  Bearded  Yalture,  Salt,  App.  p.  xlL 
G,  meridionaUs,  Keys  and  Bias.  Wirb.  Europas,  p.  xxyu. — Riipp. 

Syst.  Ueb^s.  No.  1,  t.  1.— Ferret  et  GalL  No.  1. 
G,  barbcUuSy  Des  Mors,  p.  43. 
G,  barbatHS  meridionalis,  HeagL  Om.  N.  0.  Afir.  No.  9. 

Iris  whitish  yellow,  outer  circle  of  opaque  scarlet, 
feet  pale  olive. 

The  Abyssinian  Lammergeyer  was  occasionally  but  not 
frequently  seen  at  the  base  of  the  hills.  It  was  more 
common  in  the  passes,  and  abundant  on  the  highlands, 
especially  at  all  camps  in  the  neighbourhood  of  crags. 
In  the  summer  months  perhaps  all  keep  to  the  highlands, 
as  none  were  seen  on  the  Anseba  in  July  or  August 

Abundant  as  these  birds  were  around  Senaf^  and 
other  places,  I  never  saw  them  descend  on  carcases 
like  Vultures,  but  they  appeared  to  search  for  small 
fragments  of  offal,  &c.  around  the  camp ;  and  when  so 
doing,  their  whole  habits  and  flight  precisely  resemble 
those  of  a  Kite.  Indeed,  so  great  is  the  resemblance 
of  this  bird  to  a  Kite,  that  I  cannot  help  suspecting  that 


AVES.  299 

its  affinities  are  with  the  FalconidcB  near  the  MilvincB, 
and  not  with  the  Vulturea  The  appearance  when 
standing  on  the  ground  is  certainly  somewhat  like  that 
of  Neophron  percnopteruSy  but  this  bird  is  itself  a 
very  aberrant  member  of  the  Vulturidce.  The  flight 
of  the  Lammergeyer  is  very  fine,  equalling  that  of  a 
Vulture.^     It  keeps  much  to  precipices. 

Some  authors  consider  the  Abyssinian  Lammergeyer 
as  only  a  variety  of  the  Alpine  and  EUlmalayan  one. 
Yet  the  diflerences  between  the  two  birds  are  very 
considerable.  Gypaetos  memdionalis^  in  addition  to  its 
naked  tarsi  and  deeper  coloration,  is  very  much  smaller 
than  the  European  and  Asiatic  race,  as  the  following 
measurements  of  two  birds  shot  at  Senaf^  show.  The 
dimensions  were  taken  in  the  flesh  : — 

Toung  fenude.       Old  male, 
ft    in.  ft.   in. 

Length 37  33 

Extent  of  wings 80  74 

Length  of  closed  wing 25  23 

„        tarsus 04^  — 

„        middle  toe 04^  — 

„        beak  to  gape 0    4  — 

The  European  Lammergeyer  is  4^  ft.  long,  and  has  an 
extent  of  10  ft.  according  to  Naumann.  In  the  Hima- 
layan bird  the  male  measures  46  in.  in  length,  9  ft.  in 

1  Few,  who  have  not  been  in  the  habit  of  seeing  Valtures,  can  easily  con- 
ceive what  noble  birds  they  are  when  on  the  wing.  The  sluggish,  slovenly, 
ugly  scavenger  becomes  at  once  a  different  being,  with  the  most  graceful  and 
powerful  flight.  The  manner  in  which  they  can  fly  around  crags,  keeping 
themselves  in  swift  motion  apparently  for  an  indefinite  time,  rising,  falling, 
or  soaring  in  circles  without  moving  their  wings  or  tail,  except  to  guide 
themselves,  is  simply  marvellous. 


300  ZOOLOGY, 

extent ;  the  female  4  ft.  in  length  and  9^  in.  in  extent 
(Jerdon,  "Birds  of  India,"  p.  14). 

Bnice's  account  of  the  Abyssinian  bird  is  very  amusing. 
He  records  a  singular  act,  the  seizure  of  some  goat's 
flesh  from  his  servants  by  a  Lammergeyer. 

zg.  Milvus  migrans,  Bodd. 

M,  aier,  Riipp.  Syst.  Uebers.  No.  37,  partim. — Ferret  et  Oallinier. 

No.  13. — Lefebvre,  p.  69. 
M,  atolius  (!)  Heugl.  Cm.  N.  0.  Afr.  No.  64. 

Extremely  common  everywhere,  both  on  the  highlands 
and  lowlands.  As  I  brought  several  specimens  with 
me  from  various  localities,  I  have  certainly  not  mistaken 
M.  (Bgyptius  for  this  bird.  Von  Heuglin  is  decidedly  in 
error  in  supposing  it  to  be  only  a  winter  visitant  in 
North-eastern  Africa.  He  appears  to  doubt  its  occurrence 
in  Abyssinia.  A  similar  circumstance  occurred  to  me  to 
that  quoted  in  the  Zoological  portion  of  Lefebvre's  "Voyage 
en  Abyssinie."  I  was  skinning  a  specimen  of  Scopus 
umhretta  at  Lat  when  a  Kite  swooped  and  attempted  to 
carry  it  off,  scoring  my  fingers  in  the  attempt 

20.  M.  segy ptius,  Gm. 

Falco  (jBgyptius,  Gmelin,  Syst  Nat  i.  261. 

F.  Forshi^lii,  Ihid.  p.  263. 

F,  par(mticu8,  Latham. 

IMUvus  aitr,  Riipp.  Syst  Uebers.  No.  yiypariiiiu 

M,  parasiticus,  HeagL  Syst  Uebers.  No.  60. — Brehm,  Habesch, 

No.  14. 
M,  Forskali,  HeugL  Ora.  N.  0.  Afr.  No.  66. 

Iris  brown  ;  bill  and  legs  yellow. 
This   appeared  to  me  much  less  common   than   the 
preceding.      I   did   not   distinguish   the   two   when   in 


AVES.  301 

Abyssinia,  and  the  only  specimen  I  preserved  of  this 
Kite  was  killed  on  the  highlands  at  Lake  AshangL 
Heuglin  must  be  under  some  error  in  treating  of  this 
as  the  only  common  Kite  of  North-eastern  Africa. 
Certainly  so  far  as  my  specimens  prove,  M.  ater  was 
more  abundant  in  Abyssinia ;  and  unquestionably  it  was 
the  common  species  both  at  ZuUa  and  Senaf^,  as  I  shot 
several  specimens  at  each  locality.  I  take  the  first  name 
in  Gmelin,  and  do  not  understand  why  the  German  omi- 
thologists  prefer  the  less  well-known  term  M.  Forskali. 

21.  Circus  fieruginosus,  Linn. 

C  rufm,  Riipp.  Syst.  Uebers.  No.  45. — HengL  Syst.  Uebers.  No.  72. 
0.  ceruginoiuSy  Heagl.  Oni.  N.  0.  Afr.  No.  68. 

Occasionally  seen  on  the  highlands.  I  never  met  with 
it  in  the  peculiar  adult  livery  of  the  Indian  race. 

22.  C.  Swainsoni,  A.  Smith. 

S.  Af.  Quart.    Joum.  L  384,  1830.— S.  AX  Zool.  plates  43,  44.— 

Heugl.  Orn.  N.  0.  Afr.  No.  71. 
C.  palluliiSy  Sykes,  P.  Z.  S.  1832,  p.  80. 
Struficeps  pallvlus,  Brehm,  Habesch,  No.  17. 

23.  C.  cineraceus  (Mont.). 

C.  cinerascens,  Heugl.  Ora.  N.  0.  Afr.  No.  70. 

Both  these  species  abounded  on  the  highlands  in  the 
winter  and  spring.  I  shot  a  young  male  of  C,  cineraceus 
on  the  19  th  August  close  to  the  coast  in  Samhar,  about 
twenty-five  miles  north  of  Massowa,  This  is  early  for  a 
Harrier  to  be  found  so  far  south,  but  they  migrate  into 
India  sooner  than  most  of  the  winter  visitants. 


302  ZOOLOGY, 


Family  STRIGID^ 

24.  Bubo  lacteus,  Temm. 

jSECrix  loctea^  Temm.  PI.  ooL  4. 

Bvibo  lacteus,  Biipp.  Syst.  Ueben.  No.  53. — HeugL  Orn.  N.  0.  Afr. 

No.  79. 
B.  cinerascens,  Brehm,  Habesch,  No.  19. 

Iris  pale  brown ;  beak^  cere,  and  feet  very  pale  hom- 
colour,  almost  whitish  ;  claws  rather  darker. 

This  bird  I  did  not  meet  with  on  the  highlands.  A 
specimen  was  shot  by  Captain  Sturt,  in  the  pass  below 
Senafi^,  and  I  obtained  two  in  the  Anseba  valley  and  saw 
others.  They  perch  on  high  trees  during  the  day,  and 
come  out  at  dusk. 

Despite  their  large  size,  these  Owls  are  very  insecti- 
vorous. In  the  stomachs  of  those  killed  I  found  only 
Crickets  or  Locusts  and  large  larvae,  apparently  of  cole- 
optera.  According  to  Von  Heuglin  ("  Om.  N.  0.  Afr." 
p.  114),  Bubo  lacteus  kills  small  mammals  and  birds 
and  even  Guinea-fowls,  which  it  attacks  at  night  when 
roosting  on  trees! 

25.  Bubo  cinerascens,  Qn^r. 

Oa^rin,  Bey.  ZooL  1843,  p.  321. 

B.  cineraceus,  Ferret  et  GalL  Voya^  iiL  181 ;  AUas,  pL  2. 
Otu8  cinera4ieus,  Lefeb.  p.  74 ;  Atlas,  pi.  iy. 

Bfiho  maculoius  (?  Brehm,  Habesch,  No.  20). — Heu£^  Om.  N.  0. 
Afr.  No.  80. 

Iris  brown* 

I  only  once  met  with  this  bird.  Lieutenant  St.  John 
noticed  that  a  pair  of  Owls  inhabited  a  rocky  cliff  near 


AFE8.  303 

AntaJo^  and  on  going  to  the  spot  I  was  fortunate 
enough  to  secure  one,  as  both  flew  at  the  sound  of 
shot  fired  at  a  Hyrax.  The  skin  is  that  of  a  female, 
and  measures,  whole  length  18f  in.,  wing  13j^,  tail  7, 
tarsus  2^. 

Both  Mr.  J.  H.  Gumey  and  Mons.  Jules  Verreaux 
consider  the  present  species  quite  distinct  from  Bvbo 
'maculosics  of  South  Africa.  Mr.  Gumey  informs  me  that 
he  has  seen  both  species  alive,  and  that  the  distinction  is 
generic. 

26.  Athene  perlata,  YieilL 

A,  punUa,  Biipp.  Syst.  Uebera.  No.  51. — Ferret  et  GalL  No.  20. 
A.  occipitalis,  etpugiUa,  HeugL  Syst.  Uebera.  No.  81  and  82. 
Nodua  perlaia,  HeagL  Om.  N.  0.  Afir.  No.  85. 

Iris  yellow. 

This  little  Owl  was  first  seen  and  a  specimen 
shot  at  Mayen  in  the  Senaf^  pass,  at  about  3,500 
feet.  It  was  not  rare  on  the  Anseba.  I  only  saw  it 
amongst  tiees.  It  is  somewhat  diurnal  in  its  habits,  and 
is  not  unfrequently  seen  flying  from  tree  to  tree  in  the 
evening,  some  time  before  sunset.  In  the  stomach  of  one 
specimen  I  found  remains  of  lizards. 

27.  Scops  senegalensis,  Swainson. 

Birds  W.  Afr.  I  127. 

Scops  zorca  africana,  HeugL  Om.  N.  0.  Afr.  No.  82a. 

Occasionally  heard  and  seen  in  the  Anseba  valley.  I 
found  coleoptera  in  the  stomach. 


304  ZOOLOGY. 

Order  PSITTACL 
Family    PSITTACIDiSL 
q8.  Pionus  Meyeri,  Kiipp. 

Psittactu  Meyeriy  Riipp.  Atlag^p.  16,  Taf.  11. — ^Lefeb.  p.  128. 
Pionus  Meyeriy  Biipp.  Syst   Uebera.  No.   330. — Ferret  et  Gall. 

No.  24. 
Pionius  Meyeri,  Finsch,  Papageien,  iL  494. 

Iris  pale  orange,  greenish  near  the  pupil ;  legs  and  cere 
dusky,  beak  blackish.  Usually  met  with  in  July  and 
August  in  small  parties  of  three  to  six  individuals,  less 
often  in  larger  flocks.  The  cry  is  far  less  shrill  than  that 
of  Palceomis. 

This  bird  was  never  seen  on  the  highlands.  It  was 
abundant  near  Kokai,  in  the  upper  Lebka  valley,  and 
somewhat  less  so  on  the  Anseba.  It  was  not  seen  much 
below  4,000  feet  of  elevation. 

Most  birds  have  more  or  less  yellow  on  the  crown,  the 
amount  however  varying.  EUppeirs  figure  is  that  of  a 
bird  without  any.  This  is  far  from  uncommon,  but  still 
it  is  rather  an  exception.  I  could  not  find  out  whether 
the  diflerence  is  due  to  age. 

29.  Psittacula  Tarantse,  Stanley. 

PsiUacus  Taranta,  Stanley,  Salt's  TraveLs,  Append,  p.  lii— Lefeb 

p.  128. 
Pnttaeula  TarantcB,  Riipp.  Syst.  Uebers.  No.  334.— Ferret  et  Gall. 

No.  26.— Finsch,  Papageien,  ii.  634. 

The  male  alone  has  a  red  head. 
This  bird  was  never  seen  below  7,000  feet  above  the 
sea,  and  appeared  to  be  very  locally  distributed,  aboimding 


AFES.  305 

chiefly  about  Kolqual  and  Dahro  trees  in  pairs  and  small 
parties.  It  was  especially  common  near  Adigrat  My 
collector  obtained  in  April  an  egg,  which  he  assured  me 
belonged  to  this  bird,  from  a  hole  in  a  tree,  and  there  can 
be  little  doubt  of  its  having  been  correctly  identified. 

30.  Palaeomis  torquatus  (Bodd.). 

P.  cubiculwris,  Hasselquist. — Biipp.  Syst.  Uebers.  No.  336. — ^Ferret 

et  Gal.  No.  27. 
Ptittoieus  torquatuSy  Lefebvre,  p.  128. 
Fakeomis  torquatus,  Brehm,  Habesch,   No.  110. — Finsch,  Papa- 

geien,  iL  17. 

Never  seen  by  me  on  highlands,  but  tolerably  common 
in  the  Anseba  valley  at  an  elevation  of  from  4,000  to 
5,000  feet  The  birds  obtained  by  me  had  longer  bills 
than  any  of  the  Indian  specimens  which  I  compared  with 
them,  but  Dr.  Finsch,  from  the  comparison  of  a  large 
series  of  African  and  Asiatic  specimens,  has  decided  that 
there  is  no  constant  difference  between  the  two  races. 


Order  INSESSOBES. 

Sub-Order  PIC^ 
Family  PICID^ 

31.  Picus  (Campeiliera)  nubicus,  Gm. 

Gmelin,  Syst.  Nat.  p.  4.39. 

P.  tethiopicuSy  Hempr  and  Ehr.,  Sjrmb.  Phys.,.note. 

Dendramus  athiopicus,  Riipp.  Syst.  Uebers.  p.  90,  t.  36. 

Beak  dusky,  paler  beneath.  Legs  greenish  horny.  Iris 
in  the  young  bird  pearl  grey,  in  the  adult  pink.  The 
young  bird  has  the  head  unspotted  black ;  in  the  adult  it 

X 


306  ZOO LOO r 

is  spotted  as  in  Riippell's  figure.  In  one  instance  I  killed 
three  birds  out  of  four  which  were  together,  and  doubt- 
less belonged  to  one  family.  Of  these,  two  had  the  crown 
spotted,  one  having  a  grey,  the  other  a  pink  iris ;  the 
third  had  the  crown  unspotted  and  a  grey  iria 

This  Woodpecker  was  only  met  with  between  3,000 
and  6,000  feet  above  the  sea  in  the  pass  leading  to 
Senaf^,  and  again,  to  the  north,  in  the  Lebka  and 
Anseba  valleys.  I  never  saw  any  Woodpecker  on  the 
highlands.  In  July,  on  the  Anseba,  this  was  the  only 
species  seen.  It  was  abundant  in  the  thickets  along  the 
banks  of  the  stream.  The  stomach  in  some  birds  which 
I  shot  was  filled  with  the  ova  of  insects. 

32.  Pious  {DendrobaUs)  Hemprichii,  £hr. 

Pietu  Hemprichiif  Ehrenberg,  Sjmb.  Phys. 

DendrcbaUi  Hemprichii,  Riipp.  Syst.  Uebers.  p.  S8,  t.  35. 

Picus  ahyitinicus,  Ferret  et  Gkd.  No.  40. 

P.  Hemprichii,  Lefebvre,  p.  135. 

Dendropicui  Hemprichii,  firehm,  Habeeoh,  No.  113. 

This  bird  keeps  to  a  lower  level  than  the  last  species, 
but  is  certainly  rare,  and  in  the  hot  season,  I  think, 
entirely  wanting  in  the  plains  near  the  sea,  where  how- 
ever Brehm  records  having  met  with  it  in  the  month  of 
April.^  I  met  with  it  near  Undul  Wells,  in  the  pass 
below  Senaf^  and  again  in  the  Lebka  valley  at  an  ele- 
vation of  from  about  3,000  to  4,000  feet. 

1  At  the  time  of  the  spring  rains  many  birds  appear  to  migrate  from 
the  mountains  to  the  plains  of  Samhar,  which  are  not  found  there  at 
other  times. 


AVES.  307 


Family  INDICATORIDiEL 

33*  Indicator  Sparrmanni,  Sieph. 

(7)  Moroo  or  Bee  Caokoo,  Bracers  TravelB^  App.  p.  178. 

Indicator  Sparrmanni,  Steph.,  Shaw's  Zoology,  Birds,  ix.  138. 

7.   aUnrogtrii,  Temm.   PL  eoL  367. — ^Fenet  et  GaL  No.  33. — 

Lefebyre,  p.  139. 
J.  flaviseapulaius,  Biipp.  Nea.  WirK  p.  60. 
7.  archipdagteuSy  Biipp.   Syst  Uebers.   No.   349. —  HengL   Systw 

Uebeis.  No.  49L 
7.  paU%d%radr%8y  Heugl.  Joum.  f.  Om.  1864,  p.  265. 

BiU  white,  iris  brown. 

I  am  indebted  for  a  specimen  of  this  rare  species  to 
Mr.  Jesse,  who  killed  it  on  the  Anseba,  and,  on  one 
occasion,  actuaUy  shot  a  specimen  calling  upon  a  tree  in 
which  was  a  Bees'  nest 

The  bird  is  a  male,  and  measures,  wing  4'4,  tail  3 '25, 
whole  length  about  7  inches. 

34.  I.  minor,  Stepb. 

Stephens,  Shaw's  Zoolo^,  ix.  p.  140, 

7.  diadematusy  Biipp.  Neu.  Wirb.  p.  61. 

7.  minor,  Biipp.  Syst.  Uebers.  No.  350.— Brehm,  Habesch,  No.  114. 

?  7.  major,  Lefebyre,  p.  138. 

Iris  brown,  beak  and  legs  dusky,  nostrils  subtubular 
with  raised  margins  as  in  cuckoos. 

I  shot  two  specimens  of  this  bird  in  the  thickets  on 
the  banks  of  the  Anseba,  where,  however,  it  appeared 
to  be  far  from  common.  In  one  instance  the  Indicator 
was  in  the  act  of  chasing  and  fighting  with  a  Wood- 
pecker, Picus  cethiopicus ;  neither  bird,  however,  uttering 
any  noise.  The  Indicator,  when  it  settled,  clung  to  a 
stem  like  a  Woodpecker,  but  I  did  not  see  it  climb  in 
the  same  way. 

X  2 


308  ZOOLOGY. 

The  stomach  is  not  muscular  but  quite  thin,  and  in 
both  specimens  of  this  species,  as  well  as  in  those  of 
I.  Sparrmanniy  filled  with  a- mass  of  wax,  from  which, 
however,  all  the  scent  of  honey  was  gone.  In  one 
instance,  at  least,  fragments  of  hymenopterous  insects  in 
considerable  quantities  were  mixed  with  the  wax :  they 
looked,  however,  more  like  portions  of  ants  than  of  bees. 
There  can  apparently  be  no  doubt  that  both  species  of 
Indicator,  feed  on  honey  and  attack  Bees'  nests,  probably 
feeding  also  on  the  Bees  themselves.  The  native  stories 
of  their  leading  people  to  Bees'  nests  are  well  known,  but 
in  Northern  Abyssinia  at  least  it  is  asserted  that  these 
birds  do  not  alwa,ys  conduct  the  honey-seeker  to  the  same 
goal  At  times,  it  is  said,  they  lead  to  a  dead  body  ;  at 
others  to  a  Lion  or  a  Leopard.  Tales  told  of  animals  by 
uncivilized  ra^es  must,  however,  in  all  cases  be  received 
with  great  caution ;  and  in  the  instance  of  a  race  like  the 
Abyssinians,  who  a^re  not  hunters  in  general,  imagination 
probably  plays  a  far  m,ore  important  part  than  observa- 
tion, precisely  as  am,ongst  all  the  rac^  of  India,  except 
the  few  whose  life  is  spent  to  a  great  extent  in  hunting 
for  wild  animals  and  forest  products  for  food.  That  by 
observing  and  following  an  Indicator  Bees'  nestB  may  be 
found  is  easily  understood,  and  it  is  not  necessary  to 
conclude  that  the  bird  shows  the  road  intentionally. 

The  sternum  of  Indicator  does  not  much  resemble 
that  of  a  Cuckoo.  It  is  more  like  that  of  a  Woodpecker, 
but  still  more  similar  to  that  pf  a  Barbet  or  a  Coly.  There 
is  a  deep  double  notch  at  each  side  behind,  the  two 
portions  separated  by  an  oblique  rib,  the  inner  being 


AVES.  309 

the  deeper.  The  whole  bone  is  long  and  flat,  and 
rather  narrow,  becoming  gradually  broader  behind ;  the 
ridge  low  with  a  straight  edge.  Judging  from  the 
sternum,  I  should  certainly  be  disposed  to  place  Indi- 
cator with  the  Barbets  or  Colies  rather  than  with  the 
Cuckoos.  It  is  evidently  a  bird  of  weak  flight. 
.  It  is  not  clear  to  what  bird  Bruce  refers  under  the 
name  of  the  Moroc  or  Bee  Cuckoo.  His  figure  resembles 
Oxylophus  or  Chrysococcyx ;  and  if,  as  he  states,  it  is 
of  the  natural  size,  it  should  be  the  latter:  but  the 
description  of  the  colouring  by  no  means  accords,  and 
agrees  rather  with  that  of  Indicator  Sparmuxnni^ 
though  diflering  in  several  particulars,  some  of  which 
must  certainly  be  erroneous — such  as  the  statement  that 
it  has  only  three  toes.  He  relates  concerning  it  the 
usual  stories  told  of  Indicator;  which,  however,  he  treats 
as  fables :  his  own  account  of  the  habits  seems  to  agree 
best  with  those  of  a  Merops.  It  is  by  no  means 
improbable  that  the  drawing  was  made  from  Oxyhphua 
glandariits  or  0.  jacobinus,  and  the  description  drawn 
up  from  memory. 

Family  CAPITONIDiE. 

35.  Pogonorhynchus  abyssinicus  (Latham). 

Phytotoma  abysdnicay  Latham,  Ind.  Om.  Sapp.  xlix. 
Bucco  Saltii,  Stanley,  Salt^s  Journey,  Appendix,  p.  liv. 
Fogonias  Bruceiy  Riipp.  Neu.  Wirb.  p.  50,  t  20,  fig.  1. 
Laimodon  Bruceiy  Riipp.  Syst.  Uebers.  No.  338. 
Pogonias  gnisso  bcditOy  Bruce  ?    Ferret  et  GraL  No.  38. 
P.  rubrifronSy  Lefebvre,  p.  128. 
Pogonias  ScUtiiy  Brehm,  Habesch,  No.  111. 

I  obtained  one  specimen  of  this  bird  close  to  the  camp 
at  Dongolo,  on  Dahro  {Ficns  dahero)^  but  I  never  saw  it 


310  ZOOLOGY. 

on  the  higher  portions  of  the  plateau  above  7,000  feet; 
In  July  pairs  were  occasionally  seen  on  high  trees  making 
a  great  noise  and  probably  engaged  in  breeding. 

36.  P.  undatus,  Eiipp. 

Fogoniaa  undatus,  Biipp.  Nen.  Wirb.  p.  52,  t.  20,  fig.  2. — Ferret  et 

GaL  No.  36. 
Laimodan  widatuSy  Riipp.  Syst  Uebera.  No.  338. — Heog^  Syst. 

Uebers.  No.  477. 
Bucco  undatus,  Lefebvre,  p.  130. 
B.  undulcUuSf  Des  Muis  et  Pr^yost,  Lefebyie,  Atlas,  pi.  10,  fig.  2. 

Iris  pale  greyish  yellow,  bill  black,  legs  dusky.  Not 
rare  about  Adigrat  and  Senaf^. 

Close  to  the  shrine  of  St.  Eomanos  I  shot  a  specimen 
which  was  hanging  head  downwards  from  a  branch, 
feeding  upon  fruit 

37.  P.  melanocephalus,  Eiipp. 

Fogonias  melanocephalus,  Biipp.  Atlas,  p.  41,  t.  28,  fig.  a. 
Bxicco  hifrenatuSy  Hemp,  and  Ehr.,  Symb.  Phys.,  t.  yiii  figs.  1,  2. 
LoMnodon  melanocephaliu,  Biipp.  Syst.  Uebers.  No.  336. 

Iris  brown,  beak  and  legs  black. 

Only  seen  in  the  neighbourhood  of  the  coast  I 
obtained  one  specimen  at  Komayli  and  two  others  near 
Ailat  Apparently  the  three  species  of  Pogonorhynchus 
are  thus  to  a  certain  extent  characteristic  of  the  three 
lower  subdivisions  of  the  Abyssinian  fauna ;  P.  melano- 
cephalus being  met  with  in  the  tropical,  P.  Saltii  in  the 
subtropical,  and  P.  undatvs  in  the  temperate  region. 

In  the  stomach  of  P.  melanocephalus  were  many 
coleoptera.  All  the  Abjrssinian  Barbets  appear  much 
more  insectivorous  than  their  Indian  representatives. 
The  same  is  the  case  with  the  Hornbills. 


AFRS.  311 

38.  Bdrbatula  pusilla,  Dumont. 

Bucco  pusiUuBy  Dum.,  Diet  des  Sc.  Nat  iv.  p.  50. 
B.  ehryioeomusy  Riipp.  Syst  Ueben.  No.  341. 
B.  ekry$oeomu8y  Ferret  et  OaL  No.  36. 

This  bird  was  only  met  with  in  the  subtropical  region. 
It  was  far  firom  rare  about  Mayen  in  the  Senaf^  pass, 
and  common  on  the  Anseba.  It  has  a  peculiar  simple 
note,  recalling  that  of  the  well-known  "Coppersmith '* 
of  India  {Megalcema  indica)^  but  rather  less  metallic 
and  differently  uttered;  for  while  the  call  of  the  latter 
is  single  with  regular  brief  intervals,  and  repeated  thus 
for  a  considerable  time,  that  of  the  former  is  double 
or  treble  followed  by  a  pause. 

Von  Heuglin  describes  another  species,  B.  uropygialis, 
from  the  Anseba,  and  states  that  he  found  it  common 
there.  It  is  evidently  not  the  same  as  that  which  Mr. 
Jesse  and  I  obtained,  for  it  has  the  back  sulphur  yellow 
in  the  centre  and  the  rump  orange,  besides  many  other 
distinctions. 

39.  Trachyphonus  margaritatus  (Riipp.). 

Buceo  margarUcUtUy  Biipp.  Atlas,  t  20. 

TamuUia  erythropygos,  Hemp,  and  Ehr.,  Symb.  Phys.,  t  viL 

Tra4:hyphonus  margaritatusy  Biipp.  Syst.  Uebera.  No.  342. — Brehm 

Habesch,  No.  112. 
Micropogon  margaritcUus,  Ferret  et  Gal.  No.  34. 
Af.  margaritaceuSy  Lefebvre,  p.  132. 

Iris  brownish  grey,  bill  pale  reddish  brown,  legs 
greenish  homy. 

Common  from  the  sea-coast  to  about  3,000  feet  in  the 
Senaf^  pass,  and  very  abundant  on  the  Anseba.  I  believe 
that  I  heard  it  at  about  7,000  feet  on  the  highlands,  but 


312  ZOOWGY, 

I  never  heard   or  saw  it  above  that  elevation.     It   is 
insectivorous  to  a  considerable  extent. 

Its  note  is  quite  different  from  that  of  the  typical 
Barbets,  being  a  rapid  repetition  of  two  notes,  and  so 
similar  to  that  of  the  common  grey  Francolin  of  India^ 
F.  {Ortygiornis)  pondicerianus,  that  for  a  long  time  I 
supposed  it  to  be  the  call  of  a  Partridge.  In  June,  July, 
and  August  it  was  constantly  heard.  This  Barbet  keeps 
much  to  thick  low  bushes. 

Family  CUCULIDiE, 

40.  Cuculus  canorus,  L. 

I  heard  a  Cuckoo  on  the  1 7th  May  at  Agula,  about 
half-way  between  Adigrat  and  Antalo.  This,  however, 
scarcely  proves  that  the  bird  breeds  in  Abyssinia,  for  I 
have  heard  the  woods  resounding  with  the  well-known 
note  in  the  centre  of  India,  near  Sironcha  on  the 
Godavery,  at  the  same  season.  In  the  Anseba  valley 
Cuckoos  made  their  appearance  at  the  end  of  July,  but 
they  were  not  then  calling.  I  shot  one,  a  male  and 
apparently  a  young  bird,  on  the  4th  August; 

41.  Oxylophus  afer  (Leach). 

Cuculus  afer.  Leach,  Zool.  Misc.  i.  p.  72. 

C.  edolio,  Leyaill.  var.  Salt,  No.  12^  A  pp.  p.  xlVi. 

O.  cUer,  Riipp.  Syst.  Uebers.  No.  356. 

0.  LevaUlantii,  Ferr.  et  Gal.  No.  29. 

Iris  brown,  bill  black,  legs  dusky  grey. 

This  bird  was  occasionally  seen,  and  four  specimens 
in  beautiful  plumage  shot,  in  the  Upper  Lebka  and  the 
Anseba   valleys   in   July.      They   were  then   evidently 


AFES.  313 

breeding,  being  in  pairs  and  very  noisy  ;  and  a  hen  shot 
on  the  11th  of  July  contained  an  egg  fuDy  matured  and 
ready  for  laying.     It  was  bluish  green  in  colour. 

The  usual  cry  is  a  harsh  chatter  quickly  repeated,  uot 
unlike  a  Woodpecker's.  The  bird  has  also  a  liquid 
cuculine  note.  When  sitting  on  a  branch  it  often  raises 
its  crest,  which  when  raised  looks  as  if  it  were  double. 

Measurements  of  a  male : — Whole  length  15  inches, 
wing  7,  tail  8*75,  crest  1*5. 

42.  O.  jacobinus  (Bodd.). 

O.  mdanoleiLCOs,  Gmelin. 

O.  serraius,  Heugl.  Syst.  Uebers.  No.  500. 

Very  rare  in  the  Ansel)a  valley,  and  not  seen  else- 
where. I  shot  a  single  male  specimen,  and  Mr.  Jesse 
obtained  another.  I  can  see  no  distinction  between  the 
bird  I  killed  and  a  series  of  Indian  specimens  with 
which  I  compared  it. 

The  specimen  shot  measures  : — Whole  length  12*5  in., 
wing  5*75,  tail  7,  crest  1*3. 

43.  Chrysococcyx  cupreus  (Bodd.). 

Cuculus  auratus,  Gmelin. 

Chrysococcyx  auraJtus^  Riipp.  Syst  Uebers.  No.  359. — Heugl.  Syst. 
Uebers.  No.  505. 

Iris  and  orbit  scarlet,  bill  dusky  above,  pale  horny 
beneath,  legs  dusky. 

Rather  scarce  in  the  Anseba  valley  ;  no  specimens  of 
the  genus  seen  by  me  on  the  highlands.  Found  amongst 
bushes,  not  being  shy.  It  was  probably  breeding  in 
July,  for  on  one  occasion  two  were  found  sitting  together  ; 
and  both  in  that  instance  and  a  previous  one  the  birds 


Tvr^.' 


314  ZOOLOGY. 

were  singing  most  musically,  the  note  being  the  sweetest 
and  most  varied  I  ever  heard  from  a  Cuckoo.  The  bird's 
position  when  singing  is  peculiar ;  it  stoops  on  the  branch 
and  partly  spreads  its  winga  In  the  stomach  of  one 
specimen  examined  were  various  insects,  including  ants ; 
caterpillars  and  other  larvae  predominating.  Wing  4  ft. 
6  in.,  tail  3'4  in. 

44.  C.  Klaasi  (Cay.). 

VieilL  Enc.  Meth.  p.  1333.— Riipp.  Syst  Uebere.  No.  367. 

Iris  brown,  orbit  very  pale  green,  beak  dusky,  legs 
pale  brownish  olive. 

Only  a  single  specimen  obtained,  which  was  shot  on 
a  bush  in  the  Lebka  valley  at  about  4,000  ft.  elevation. 
It  is  a  male,  and  immature,  having  rufous  bands  on  the 
feathers.     Wing  37  in.,  tail  2*7  in. 

45*  Centropus  monachus,  Riipp. 

Kea.  Wiib.  p.  67,  t  21,  fig.  2.— Syst  Uebere.  No.  362.— Ferr.  et 

GaL  No.  30. 
C  Mvie^a^evMM,  Salt,  No.  13,  App.  p.  xlvL 

Iris  red. 

This  bird  is  peculiar  to  the  highlands,  and  was  never 
noticed  much  below  7,000  feet,  nor  above  8,500.  Its  range 
may  be  greater,  but  it  is  probably  confined  to  the  tem- 
perate region.  It  was  met  with  in  thick  bushes,  often 
on  the  banks  of  streams,  and  with  the  skulking,  hiding 
habits  of  the  genus,  making  its  way  upon  the  ground 
through  thick  bush.  The  flight  and  habits  are  very 
similar  to  those  of  the  well-known  "  Crow  Pheasant "  of 
India   (C    ru/ipennis).     Its  cry  also   is   similar,   very 


AFBS.  313 

quickly  uttered — who-who-who-who-wftwftwftwftwa.  A 
male  measures : — Whole  length  15*5  inches,  wing  7, 
tail  9,  length  of  hind  claw  0'8. 

46.  C.  superciliosus,  Elipp. 

Biipp.  Neu.  Wirb.  p.  66,  t  21,  fig.  1 ;  Syst  Uebers.  No.  363.— 
FezT.  et  OaL  No.  31. — ^Brehm,  Habesch,  No.  116. 

Iris  deep  blood-red,  legs  bluish  grey,  bill  black. 

I  first  shot  this  bird  at  Ailat ;  it  was  abimdant  at  Ain 
on  the  Lebka,  and  occurred  occasionally  in  the  Anseba 
valley.  It  was  never  seen  in  Abyssinia  proper.  The 
habits  are  similar  to  those  of  (7.  monachus,  but  the  cry 
seemed  to  me  rather  different,  somewhat  resembling 
the  coo  of  a  dove  uttered  shortly,  and  quickly  repeated 
several  times. 

Several  German  naturalists,  amongst  whom  are  Drs. 
Hardaub  and  Finsch,  consider  this  bird  the  young  of 
C.  monachus.  This  is  certainly  an  error,  and  one  into 
which  no  naturalist  acquainted  with  the  birds  in  the 
wild  state  could  have  fallen.  Both  species  were  founded 
on  fully  adult  specimens,  and  not  only  are  they  quite 
different  in  plumage,  but  their  habitat  is  entirely  dis- 
tinct, C.  monachus  being  only  foimd  in  the  temperate 
region  of  Abyssinia,  while  C.  superciliosus  is  equally 
confined  to  the  tropical  and  subtropical  parts  of  the- 
country.  I  saw  numerous  specimens  of  both  and  shot 
several     The  measurements  of  C.  superciliosus  are  : — 

Whole  length.  Wing.       TaiL    Hind  claw, 
in.  in.  in.  in. 

Male 1476        68         8  0*86 

Female 14-75        6*2        8*2        066 


316  ZOOLOGY, 

Family  MUSOPHAGID^. 

47.  Turacus  leucotis  (Kiipp.). 

Corythaix  leucotis^  Riipp.  Neu.  Wirb.  p.  8,  t.  3. 
Turacus  leucotis,    Riipp.   Syst.   Uebere.   No.  326. — Ferr.  et  GaL 
No.  45. 

Iris  brown,  bill  scarlet,  with  a  green  patch  at  the  base 
above  the  nostrils,  legs  dusky.  Around  the  eye  is  a 
ring  of  small  coral-red  papilte,  a  few  of  the  uppermost 
larger  than  the  rest  and  paler  in  colour. 

This  lovely  bird  abounds  in  the  subtropical  and  is 
offcen  met  with  in  the  temperate  region  up  to  8,000  feet : 
I  never  saw  it  below  4,000  feet.  It  was  common  in  the 
passes  from  Undul  Wells  nearly  to  Senaf^.  It  also 
aboimded  in  the  Anseba  valley.  It  keeps  much  to  high 
trees,  but  can  climb  well  and  quickly  amongst  rocks. 
It  is  very  often  seen  on  junipers — indeed  I  scarcely  ever 
saw  it  in  the  temperate  region  away  from  them ;  and  in 
March  I  found  it  and  most  other  fruit-eating  birds,  such 
as  Amydrxis  and  Pycnonotus^  feeding  upon  the  berries. 

48,  Chizaerhis  zonura  (Riipp.). 

Biipp.  Neu.  Wirb.  p.  0,  t  4 ;  Syst.  Uebeis.  No.  327. 

Iris  olive  brown,  beak  greenish  yellow. 

This  dull-coloured  Plantain-eater  is  very  rare  in  all 
parts  of  the  Abyssinian  highlands  traversed  by  the 
army.  I  only  once  saw  it  there  :  this  was  in  the  thickly 
wooded  valley  of  Meshek,  south  of  Antalo.  I  was  sur- 
prised at  hearing  a  peculiar  harsh  cry  quite  unknown 
to  me,  and  on  looking  up  saw  the  bird  on  a  branch 
just  above  my  head. 


AFES.  317 

111  the  Upper  Lebka  and  Anaeba  valleys  this  Okizaerhis 
abounds,  and  its  excessively  loud  harsh  call  is  constantly 
heard.  Like  other  Plantain-eaters,  it  appears  to  live 
solely  on  fruit     It  keeps  mostly  to  high  trees. 

Family  COLIID^ 

49.  Colius  leucotis,  Eiipp. 

Mils*  Senokenb^rg.  yol.  iiL  t  2^  fl^  1;  Sys^  U^ben.  No.  319.— Ferr. 

et  Gal.  No.  48. 
C.  itriatus,  Salt,  No.  44,  App.  p.  xlriii. 
(7.  Uuconotus,  Brehm,  Hab^b,  p.  219,  No^  103. 

Iris  greyish  blue  (pal^  cobalt)^  legs  dull  carmine,  re- 
sembling those  of  a  pigeon  in  colour,  g^^eater  portion  of 
upper  mandible  and  base  of  lo.wer  dusky,  centre  of 
upper  and  greater  portion  pf  lower  fleshtco^oure^ 

Inhabits  the  teipaperate  region  pf  Abyssinjai,  occasion,- 
ally  descending  to  about  4,000  or  5,000  feet  above  thpi 
sea.  It  lives  in  small  flocks  (?^  families),  which^  keep, 
much  in  thick  bushes  and  feed  on  fruit.  They  fly 
from  bush  to  bush  with  a  moderately  rapid,  very  straight 
flight,  and  evidently  have  no  great  power  of  wing,  the 
different  individuals  following  each  othei;  in  the  manner 
of  Crateropi  or  Malacoc^ciy  which  they  also  somewhat 
resemble  in  their  flight — so  much  so,  that  I  took  the  first 
flocks  I  saw  for  some  very  long-tailed  members  of  the 
CraterojndcB. 

I  saw  this  bird  frequently  in  Tigrd,  and  a  few  speci- 
mens occurred  in  the  Anseba  valley ;  but  it  w^as  less 
abimdant  there  than  the  next  species.  On  July  16th 
I  found  a  nest  with  two  very  young  birds.  It  was  a 
small  flat  platform  of  sticks  in  a  thorny  bush. 


318  ZOOLOGY, 

The  toes  of  Colitis  are  very  different  from  those  in 
other  Scansorial  Insessores,  and  even  fix)m  those  of  the 
Plantain-eaters.  Both  the  outer  toes  are  reversible  to  a 
certain  extent,  but  the  third  toe  appears  very  rarely  to 
be  reversed,  and  even  the  fourth  is  quite  as  frequently 
turned  forward  as  backwards. 

As  a  species  Coitus  hucotis  is  barely  separable  from 
the  South  African  C  sUnatus,  Gm.  The  whitish  ear- 
coverts  are  often  ill-marked ;  and  although  the  general 
tinge  in  the  Abyssinian  bird  is  slightly  more  rufous,  the 
difference  is  very  trifling.  The  principal  distinctions  are, 
that  the  wing  in  C  hucotis  is  about  ^inch  and  the  tail 
1  inch  longer  than  in  C  striatus,  that  the  former  is  more 
distinctly  striated  at  the  back  of  the  neck  than  the 
latter,  and  that  it  has  the  rump  also  transversely  striated 
t^ith  rufous  lines,  which  is  not  the  case  in  C  striatus. 
Also  in  C.  leucotis  the  crest  is  situated  further  back,  and 
the  earthy-brown  feathers  of  the  forehead  occupy  a  larger 
space  than  in  C.  striatus.  The  feet  in  the  latter,  too, 
appear  to  be  dark  brown. 

50.  Colius  macrourus,  L. 

C.  BenegaleMWy  Gm.  Syst  Nat  p.  842. — Riipp.  Syst  Ueben.  No. 
318. — Ferr.  et  Gal.  No.  47.— Lefebvie,  p.  122. 

Iris  crimson  inside,  violet  passing  to  greyish  without. 
Naked  skin  around  the  eyes  and  the  lores  deep  red.  Legs 
purplish  red.  Upper  mandible  red,  except  the  tip,  which 
is  black,  as  is  the  whole  of  the  lower  mandible.  In 
young  birds  the  iris  is  brown,  biU  greenish  yellow  with 
the  tip  whitish,  skin  around  the  eyes  dusky ;  the  blue 
feathers  of  the  nape  and  the  long  tail  are  both  wanting. 


AVES.  319 

This  bird  appears  to  range  from  about  2,000  feet  or 
even  lower  to  5,000  or  6,000  feet,  entirely  replacing  the 
last  species  in  the  tropical,  and  to  a  great  extent  in  the 
subtropical  zone.  Its  habits  are  precisely  similar  to  those 
of  C.  leticotis. 

Family  CORA.CIADJE. 

51.  Coracias  abyssinica,  Bodd. 

Sherigrig  of  Brace,  yoL  y.  p.  182,  with  plate. 

C  ahysiiniGm^  Gm. — Lefebvre,  p.  79. — Bdi^.  Syst.  Ueben.  No. 

84. — ^Biehm,  Habesch,  No.  31. 
0.  hahminuMiy  HeugL  Joum.  £  Cm.  1868,  p.  319,  et  Om.  N.O. 

Ah.  No.  134. 

Iris  raw  imiber.  The  habits  are  precisely  sdmilar 
to  those  of  C  indica.  It  is  frequently  seen  sitting  on 
trees,  or  on  bare  stems,  and  thence  descending  to  the 
ground  for  insects  or  catching  them  in  the  air. 

This  bird  is  not  rare  locaUy  on  the  highlands,  but  by 
no  means  generally  distributed.  I  saw  it  occasionally 
between  Dolo  and  Antalo,  and  again  rather  more  com- 
monly about  Lake  Ashangi  (8,000  feet  above  the  sea), 
and  in  some  of  the  valleys  further  south,  but  not  on  the 
plateaux.  It  was  very  common  in  the  subtropical  region 
of  the  Anseba  and  Lebka,  and  I  saw  one  or  two  birds,  in 
August,  in  Samhar,  near  the  coast  It  is  also  found  in 
Southern  Arabia,  and  birds  are  occasionally  seen  on  the 
Red  Sea. 

52.  Coracias  pilosa,  Lath. 

Tjatham,  lod.  Omith.  Supp.  p.  xxxiL 

C.  nuchcUis,  Swainson,  B.  of  W.  Afr.  p.  110.— Ferr.  et  Gal.  iii.  p.  238. 
0.  LevaiUatUiif  Riipp.  Syst.  Uebers.  No.  85. 
0.  pUoMf  HeugL  Joum.  f.  Om,  1868,  p.  320,  et  Om.  N.  0.  Afr. 
No.  133. 

Iris  olivaceous  brown   (almost  raw  sienna),  feet  pale 


320  ZOOLOGY. 

yellowish  olive,  bill  black  externally,  inside  of  mouth 
yellowish  olive  like  the  feet. 

This  bird  is  very  much  rarer  than  the  last  on  the 
highlands.  I  only  saw  it  twice,  and  failed  to  procure  a 
specimen.  It  abounded  in  the  subtropical  region  of  the 
Anseba. 

53.  Eurystomus  afer  (Lath.). 

Corcteias  afra,  Latb.  Ind.  Dm.  p.  172. 
EurysUnnus  ortentaliSf  Biipp.  Syst.  Uebeis.  No.  82. 
Eurystomus  afer,  HeugL  Joum.  f.  Om.  1868,  p.  322,  et  Om.  N.  O. 
Afr.  No.  131. 

Iris  olivaceous,  logs  pale  yellowish  olive,  bill  deep 
yellow.  A  noisy  bird  with  a  swift  hawk-like  flight, 
•with  all  the  usual  habits  of  a  Roller,  but  keeping  rather 
more  to  high  trees  than  the  two  other  species,  and 
perhaps  rather  more  given  to  "  rolling "  from  side  to 
side  when  flying.  It  often  hawks  locusts  and  other 
insects  in  the  air. 

ff 

E.  afer  was  only  met  with  in  the  subtropical  region  of 
the  Upper  Lebka  and  Anseba  valleys.  It  there  abounded, 
in  company  with  the  other  two  Rollers. 

Family  MER0PID.4i 

54.  Merops  viridissimus.  Swains. 

Swainsou,  B.  of  W.  Afr.,  ii.  p.  82. — Lefebyre,  p.  83. 
M,  viridis,  Riipp.  Syst.  Uebere.  No.  97. 

Common  near  the  coast,  and  especially  in  the  man- 
groves on  the  shores  of  Annesley  Bay. 

This  bird,  like  the  closely  allied  M.  viridis  of  India, 
has  the  habit  of  collecting  in  considerable  numbers  and 
of  settling  on  the  ground  occasionally. 


AFES,  321 

55.^M.  apiaster,  L. 

Kiipp.  Syst.  Uebers.  No.  96. — Ferr.  et  GaL  No.  162.— Lefebyie,  p. 
83. — Brehm^  Habesch,  No.  33. 

Eare  on  the  highlands  in  spring.  Not  seen  else- 
where. 

56.  M.  superciliosuSy  L. 

Riipp.  Syst  Uebers.  No.  96. 
M.  cegyptitts  et  M,  penicWf  auct. 

I  am  indebted  for  a  single  specimen  shot  at  Adigrat 
to  Captain  Newport.  I  did  not  myself  meet  with 
this  bird. 

Compared  with  a  specimen  fix)m  Madagascar,  the 
typical  locality,  the  wing  is  f  in.  longer.  "  The  cen- 
tral tail-feathers  are  also  longer ;  but  this  is  a  character 
varying  with  the  season. 

57.  M.  nubicus,  Gm. 

Gm.  Syst.  Nat.  L  464.— Ferr.  et  GaL  No.  163.— Lefebvre,  p.  83. 
M,  atruleocephalus,  Lath. — Biipp.  Syst  Uebeis.  No.  98. 

This  fine  Bee-eater  I  only  saw  once.  A  large  number 
were  collected  about  one  spot  close  to  the  hot  spring  of 
Atzfut  on  the  shores  of  Annesley  Bay.  Mr.  Jesse  also 
met  with  it  only  once  and  in  the  same  neighbourhood. 

58.  M.  albicollis,  Vieill. 

Vieill.  Nouv.  Diet.  d'Hist  Nat.  xiv.  p.  15,  et  Enc.  Meth.  p.  393. 
M.  Cuvieri,  Licht.  Vera.  d.  Doubl.  p.  13. — Lefebvre,  p.  83. 

Iris  crimson.  In  December,  Januaiy,  and  February 
this  bird  was  not  found  in  the  coast  region,  but  with 
some  other  species  it  migrated  into  the  country  in  the 
spring,  and  abounded  throughout  Samhar  in  June  and 
July.     I  saw  it  as  far  inland  as  Rairo  in  Habab,  but  it 

Y 


322  ZOOLOGY. 

appeared   to   be   restricted   to   the    tropical    dry    coast 
region. 

M.  alhicollis  usually  keeps  to  trees,  but  I  have  seen 
it  settle  on  the  ground,  in  company  with  M.  viridis- 
simus. 

59.  M.  Lafresnayii,  Gucr. 

?  M,  erytkropteruSf  P  large  variety.  Salt,  No.  19,  App.  p.  xlvL — 
Gu6r.  Rev.  ZooL  1843,  p.  322.— Ferr.  et  GaL  Voy.  iii.  p. 
243  ;  Atlas,  pL  15.— Brehm,  Habesch,  No.  35. 

3f.  txirugatuSj  Riipp.  Syst  Uebera.  No.  100. 

M,  Lefebvrii,  Desiu.  et  Prev. — Lefebvre,  Voyage  en  Abyssinie, 
ZooL  p.  83  ;  Atlas,  pi.  5. 

Iris  crimson,  legs  brown,  bill  black.  A  common  bird 
in  the  passes  from  3,000  ft.  upwards,  and  often  seen  on 
the  highlands,  especially  after  March. 

In  the  Anseba  Valley  this  bird  was  replaced  by  the 
next  species. 

60.  M.  erythropterus,  Gm. 

Gbl  Syst.  Nat  L  p.  464, — Salt,  No.  16,  App.  p.  xlvL — Biipp.  Syst. 
Uebera.  No.  9d. — Lefebvre,  p.  83. 

Common  in  the  Anseba  Valley  in  July.  I  unfor- 
tunately only  took  a  single  specimen,  for  the  birds  were 
moulting,  and  I  mistook  them  at  the  time  for  specimens 
of  M.  Lafresnayii  in  bad  plumage. 

Family  HALCYONIDiE. 

61  •  Halcyon  semicaerulea  (Forsk). 

Forsk.  Descr.  Animaliiim,  1775,  p.  2.— Riipp.  Neu.  Wirb.  p.  68 

t  24,  fig.  1,  et  Syst.  Uebera.  No.  87.— Ferr.  et  GaL  No.  166. 
Daetlo  8emic(eruUa,  Heugl.  Om.  N.  0.  Afr.  No.  147. 

Iris  brown,  bill  and  legs  scarlet.  A  purely  insectivo- 
rous bird,  rarely  seen  near  water,  and  occasionally  foimd 


AVES.  323 

in  dry  portions  of  the  jungle.  It  occurred  about  Ailat 
and  Ain  at  the  base  of  the  hills,  and  was  also  seen  in  the 
Anseba  valley. 

Hartlaub    (Syst.    Omith.     West-Afrika's)     separates 
Swainson's  Halcyon  rufiventer  (B.  of  W.  Africa,  ii  p.- 
101)  from  this  species;  but  my  specimens  agree  very 
well  with  Swainson's  figure  and  description. 

62.  H.  senegalensis  (liim.). 

Ferr.  et  GalL  No.  171. 

H.  cancTophagOy  Heugl.  Syst  Ueben.  Na  129  (nee.  Lath.). 

Dacdo  unegalensU,  HeugL  Om.  N.  0.  Afr.  No.  148. 

Iris  brown,  upper  mandible  red,  lower  mandible  black, 
legs  purplish  black,  soles  of  feet  salmon  colour.  This  is 
also  an  insectivorous  species.  It  was  only  met  with  on 
the  Anseba,  where  it  occurred,  far  from  commonly,  in  the 
thickets  along  the  bank  of  the  stream.  It  was,  however, 
never  seen  sitting  beside  the  water,  nor  were  there  any 
fish  in  the  torrent  which  could  have  attracted  it,  and  the 
stomach  in  all  the  specimens  killed  contained  insects 
only.  This  bird  has  a  very  loud  shrill  cry,  quickly 
repeated,  not  unlike  the  alarm-note  of  a  Woodpecker. 

Apparently  this  Kingfisher  migrated  into  the  Anseba 
valley  about  the  latter  end  of  July,  as  it  was  not  seen 
there  before. 

63.  Ispidina  picta  (Eodd.). 

Alcedo  casruUa,  EuhL— Biipp.  Neu.  Wirb.  p.  70,  et  Sjst.  Ueben* 

No.  93. 
Ispidina  cyanoiis,  Brehm,  Habescb,  No.  32. 
I,  pida,  Sharpe,  Alcedinidse,  pt.  iv. 
Alcedo  pictoj  HeugL  Omith.  N.  0.  Afr.  No.  143. 

Iris  brown,  bill  and  legs  scarlet.     In  the  young  bird 

Y  2 


324  ZOOLOGY, 

the  beak  is  black  with  a  yellowish  tip,  and  the  le^ 
pale  orange. 

This  species  also  is  insectivorous.  It  was  not  very 
nnfrequent  in  the  Anseba  valley,  keeping  to  the  thickets, 
and  often  sitting  in  the  densest  bushes.  It  was  not  seen 
elsewhere. 

A  young  bird,  which  had  apparently  only  just  left  the 
nest,  was  shot  on  July  30th.  A  few  days  later,  in  the 
evening,  a  pair  of  adult  birds  were  observed  keeping 
about  one  spot,  as  if  they  had  a  nest  there.  The  nest, 
however,  could  not  be  found.  The  sppt  was  in  rather 
thick  bushes  near  a  small  ravine. 

64.  Corythornis  cyanostigma  (Rupp.). 

Alcedo  cyanostigmaj  Riipp.  Neu.  Wkb.  pw  70,  fig.  24^  f.  2^  et  Syst. 

Uebers.  No.  92.--FeiT.  et  Gal.  No.  170. 
C  crittaiay  Sharpe,  Alcedinidae,  part  vi. — Heugl.  Ora.  N.  0.  Afr. 

Nob.  140  and  142. 

Iris  brown,  legs  and  bill  coral-red.  Only  seen  on  the 
highlands,  but  not  much  above  7,000  feet.  It  w;as  most 
common  near  Agula  and  Dongolo,  but  was  nowhere 
abundant. 

The  habits  of  this  bird  are  entirely  different  from 
those  of  the  previous  species.  It  is  a  fish-eater,  and  was 
never  seen  away  from  the  streams,  on  the  banks  of  which 
it  might  be  observed  resting  on  a  branch  over  the  water, 
and  thence  dashing  down  upon  fish,  or  skimming  rapidly 
up  or  down  the  stream  precisely  like  Alcedo  ispida, 

A,  cyanostigma  of  Riippell  was  founded  on  a  young 
specimen.  The  adult  has  always  been  looked  upon  aa  the 
true  A,  cristata,  L.,  and  the  two  have  been  separated  by 


AFES.  325 

most  ornithologists.  I  learn  from  Mr.  Sharpe  that  since 
publishing  the  description  of  this  species  in  his  "  Alcedi- 
nidsB,"  he  has  become  convinced  that  the  true  A,  cristata 
of  Linnaeus  is  the  Madagascar  species,  and  that  the 
African  form  must  bear  Ruppell's  name.  This  is  to  be 
regretted,  as  the  name  is  objectionable. 

65.  Alcedo  semitorquata,  Swains. 

Swains.  Zool.  111.  t.  151. — Riipp.  Syst.  Uebers.  p.  20,  t.  7. — Sharpe 
Alcedinidaj,  pk.  v. — Heugl.  Cm.  N.  O.  Afr.  No.  139. 

I  am  indebted  to  Captain  Newport  of  the  Commissariat 
department  for  the  only  specimen  of  this  bird  which  I 
saw  in  Abyssinia.  This  was  killed  at  Adigrat  on  the 
highlands  of  Tigr6. 

66.  Ceryle  rudis,  L. 

Riipp.  Syst.  Uebers.  No.  89.— Lefebrre,  p.  8().— Ferr.  et  GaL  No. 
168.— Heugl.  Omith.  N.  0.  Afr.  No.  144. 

This  also  was  a  rare  bird  in  the  country  traversed, 
doubtless  in  consequence  of  the  few  large  streams.  I 
saw  it  on  the  Jedda  river  near  Magdala,  and  again 
near  Dongolo,  at  about  7,000  ft.  above  the  sea,  which  I 
suspect  to  be  a  higher  elevation  than  this  Kingfisher 
usually  aflfects.  I  shot  one  specimen  at  the  latter  place. 
It  differs  considerably  from  all  Indian  specimens  with 
which  I  have  compared  it,  in  the  much  less  amount  of 
black  on  the  breast  and  the  larger  quantity  on  the  sides 
of  the  abdomen.  It  has  also  a  much  shorter  bill ;  but 
this  is  a  variable  character  in  Kingfishers.  Mr.  R.  B. 
Sharpe,  however,  considers  that  there  is  no  constant 
difference  between  African  and  Asiatic  birds,  both  vary- 
ing  much. 


826  ZOOLOGY. 

Familt  BUCEROTIDiE. 
67.  Toccus  Hemprichii,  Ehr. 

TockuB  Htm/prichiiy  Ehr. 

BuceroB  {Lophoceros)  Hemprichii,  Ehr.,  Symb.  Phys.,  Aves,  1828. 
B,  limh(Uu8,  Biipp.  Neu.  Wirb.  p.  6,  t.  2,  fig.  1,  1835. 
Todhis  IvndxUus,  Riipp.  Syst.  Uebers.  No.  325.— Ferr.  et  (laL  No. 
43.--Brehm,  Habeach,  No.  107. 

Iris  yellow  brown,  clouded  with  darker  colour  near 
the  pupil,  bill  dull  red,  legs  dusky. 

A  highland  species,  but  by  no  means  very  common  in 
the  country  traversed.  It  was  most  frequently  seen 
around  Senaf^  in  small  flocks,  probably  families,  of  about 
four  to  eight  birds.  I  occasionally  met  with  it  in  the 
higher  parts  of  the  Anseba  valley,  keeping  to  the  hills. 
On  the  Wadela  plateau,  near  Bethor,  I  saw  what  appeared 
to  me  a  darker  variety,  or  distinct  race ;  but  I  had  not 
the  means  of  procuring  a  specimen  at  the  time. 

The  measurements  of  a  female  are  : — Wing  11*4  in. 
tail  10*5 ;  bill,  nostril  to  point  4*1,  gape  to  point  4'7 ; 
height  of  bill  1  "4  ;  tarsus  1  '3  :  whole  length  about  23  in. 

By  Brehm,  "  Reise  nach  Habesch,"  p.  219,  and  by 
Hartlaub,  "Syst  der  Om.  West  Afr."  p.  164,  Bitceros 
Jlemprichii  of  Ehrenberg  is  considered  a  synonym  of 
JB.  ncisutuSy  Hartlaub  especially  calling  it  the  female. 
This  is  evidently  a  mistake,  the  female  of  B.  nasutiLS 
being  the  B.  hemileucos  of  the  "Symbolse  PhysicaB." 
Ehrenberg's  descriptions  are  not  very  good,  but  the  cha- 
racters of  the  bills  alone  serve  to  show  what  the  birds 
really  are  to  which  they  refer. 

It  is  also  as  well  to  call  attention  to  the  fact  that  the 
wbgeneric  names,  Ahphius  proposed  by  Ehrenberg  in 


AVES.  327 

the  "  Symbolse  Physicse "  for  Buceros  erythrorhynchits, 
and  Lophoceros  for  B.  nasutics  and  B.  Heviprichii,  have 
priority  over  Lesson's  genus  Tockus  proposed  for  B. 
eryihrorhynchus  and  its  allies  in  1831  (Less.  "Traits 
d'Omithologie,"  p.  252). 

68.  T.  flavirostris  (RUpp.). 

Biiceros  flavirostris,  Riipp.  Neu.  Wirb.  p.  6,  t.  iL  f.  2. — Ferr.  et 

Gal.  No.  42. 
Yellow-billed  Hombill,  Latham,  Gen.  Hist.  Birds,  ii.  331. 
Tochus  flavirostris,  BUpp.  Syst.  Uebers.  No.  224. 

Iris  pale  yellow,  pupil  small,  bill  orange  yellow,  gape 
dusky,  legs  dusky.  The  note  is  a  harsh  cluck  several 
times  repeated,  especially  when  the  bird  is  disturbed. 

In  January  and  February  this  Hornbill  was  abundant 
in  the  pass  leading  to  the  highlands  from  about  2,500 
to  5,000  feet  above  the  sea.  It  was  usually  in  small 
flocks.  In  May  and  June  all  had  migrated  to  higher 
ground,  and  were  met  with,  often  singly,  about  Senafd 
at  7,000  and  8,000  feet. 

Like  its  allies,  T.  Jlavirostns  is  partly  insectivorous. 
In  the  stomach  I  found  beetles,  ants,  and  seeds.  It  often 
feeds  on  the  ground.  All  the  Abyssinian  Tocci,  like  the 
Barbets,  appear  much  more  insectivorous  than  their  Indian 
representatives.  The  bill  varies  greatly  in  size  in  the 
two  sexes,  as  it  also  does  in  the  next  two  species.  The 
measurements  are : — 


Male    .     . 

Wing. 

In. 
.     .         8 

T&IL 

in. 
9 

Bill  ft-om 
nostril  to  tip. 

n. 

2-9 

BUI  from 
gape. 

in. 

3-3 

Height  of  bill 
acnws  nostrlL 

in. 

1-36 

Female 

.     .      '7-3 

8-7 

2-2 

2-8 

115 

328  ZOOLOGY. 


69.  T.  crythrorhynchus  (Temm.). 

Buceros  erythrorhynchus,  Temm.  PL  CoL  iL  p.  59. — Feir.  et  GraL 

No.  41. 
Bueeros  {Alophitu)  erythrorhynchui,  var.  Itueopareus,  Hemp,  and 

Ehr.,  Symb.  Phya. 
Todnu  erythrorhynchusy  Riipp.   Syst  Uebers.  No.  322. — Brehm, 

Habesch,  No.  108. 

Iris  brown,  skin  in  front  of  the  eye  pale  yellow,  bill 
dull  red,  base  of  lower  mandible  dusky  beneath,  legs 
dusky. 

Except  the  colour  of  the  iris  and  bill,  and  the  shape  of 
the  latter,  there  is  very  little  to  distinguish  this  Hombill 
fix)m  the  last.  T.  erythrorhynchus,  however,  inhabits 
a  lower  zone  of  elevation  in  general  than  its  congener. 
The  two  occurred  together  about  Mayen  (3,500  feet) 
in  the  Senafd  pass,  in  January  and  February,  but  T. 
erythrorhynchus  was  also  found  in  Samhar  in  July 
and  August.  It  was  also  common  throughout  the  Lebka 
valley  and  on  the  Anseba,  but  I  never  jsaw  it  on  the 
highlands. 

In  July  and  August  these  birds  were  Very  lively. 
Their  call  is  something  like  "  whoot-whoot-whoot-whoot- 
whoot-tawut-tawut-tawut-tawut-tawut,"  increasing  in 
intensity  and  rapidity  and  rising  in  the  scale  towards 
the  end.     Measurements : — 

win<»     Toil        Bill,  nostril      Bill,  gap«      Hei^t  of  bill       ' 
wing.     lai.  j^jjp  to  tir.        ftcroMnostriL 

in.        in.  iu.  '  in.      '  in. 

Male 7*4     9  315  3*8  1 

Female      ....    7*2     8*7         245  31  095 


AVES.  329 


70.  T.  nasutus,  L. 


BtLceros  tiamtuSf  Linn.  Syst.  Nat.  12th  ed.  i.  p.  154. 

Tockus  ncuutus,  Riipp.  Syst.  Uebers.  No.  323.— Brefam,  Habesch, 

No.  106. 
Buceros  (Lophoceros)  Forakalii,  Hemp,  and  Elir.  (male) ;  B.  (Lopho- 

ctros)  hemil^icoSf  Hemp,  and  Ehr.  (female),  Symb.  Phys. 
Tochu  poicilorhynchtLs,  Lafresnaye    (female),    Rev.  ZooL    1839, 

p.  267. 

Iris  brown,  legs  dusky,  soles  and  interstices  between 
the  scales  pale.  The  bill  differs  greatly  in  the  two  sexes. 
In  the  male  it  is  much  larger  and  black,  except  one  spot 
below  the  nostril :  in  the  female  the  upper  mandible  is 
yellowish  white  towards  the  base,  except  around  the 
nostrils,  where  it  is  dusky;  the  lower  mandible  dusky 
near  the  base,  with  one  or  two  diagonal  pale  streaks ; 
both  mandibles  reddish  brown  towards  the  tip. 

Not  very  common  about  Komayli,  at  the  foot  of  the 
hills,  in  Februaiy ;  much  more  frequent  in  the  Lebka 
valley  in  July,  and  not  rare  in  the  Anseba. 

Like  the  others,  this  bird  is  often  seen  on  the  ground,^ 
and  is  highly  insectivorous ;  coleoptera,  ants,  and  seeds 
being  found  in  the  stomach.  Amongst  the  coleoptera  in 
one  instance  were  several  of  a  species  of  Molui*us.  At 
another  time  I  saw  a  bird  of  this  species  fly  after  and 
capture  a  locust  in  the  air.  The  Hornbill  was  not  very 
skilful  in  hawking,  and  the  locust  eluded  him  at  first,  bu. 
was  soon  caught. 

It  is  not  quite  clear  that  this  species  is  not  distinguish- 

^  Brehm^s  observations  on  the  Abyssinian  Hornbills  are  opposed  to  mine. 
He  says  they  rarely  descend  to  the  ground,  that  they  feed  on  fruit«,  and  that 
the  voice  of  different  species  is  similar  ('*  Reise  nach  Habesch,"  p.  357).  I 
examined  the  stomachs  in  numerous  instances. 


330  ZOOLOGY. 

able   from  that  of  South  Africa.     If  so,  it  must    bear 
Ehrenberg's  name,  B.  Forskali.     The  dimensions  are  : — 

.an 'p.n       Bill,  nostril     Bill,  gap«  Hei|^t  of  bill 

Wing.    Tall.  ^^jp  4^^p  amwsnostrlL 

in.        in.  in.  in.  in. 

Male 87     8*4  33  3*9  1-2 

Female     ....    7-6     71  2*6  34  1-05 


71.  Bucorvus  abyssinicus  (Gm.). 

Buceros  ahymnicus^  GmeL  Syst  Nat  i.  p.  358. 

Abbagumba  and  Erkoom,  Bruce,  Travels,  voL  v.  p.  169,  with  plate. 

Tragopan  ahyssinictiSy  Riipp.  Syst.  Uebere.  No.  320. 

Bucorax  abyisinicuif  Brehm,  Habesch,  No.  109. 

Iris  brown,  legs  dusky,  bill  and  casque  black.  In  the 
female  the  skin  of  the  neck  is  purple,  in  the  male  bright 
red,  in  part  at  least,  and  at  all  seasons ;  but  I  believe  the 
extent  of  the  red  differs  in  different  individuals,  perhaps 
with  age,  and  it  may  be  absent  in  the  young. 

The  Abbagumba  is  one  of  the  most  extraordinary  birds 
of  Africa.  Belonging  to  a  family  essentially  consti- 
tuted for  perching  and  climbing,  it  has  the  legs  modified 
so  as  to  enable  it  to  seek  its  food  entirely  on  the  ground. 
To  any  one  accustomed  to  the  HombiUs  of  India  and 
Malayasia  it  is  as  startling  to  see  one  of  the  family  stalk- 
ing over  the  ground  in  an  open  plain,  as  it  would  be  to 
find  plovers  feeding  in  trees. 

I  paid  some  attention  to  the  habits  and  food  of  this 
curious  bird.  It  is  almost  entirely  insectivorous  ;  in  every 
caae  in  which  I  examined  the  stomach  the  contents  were 
principally  large  beetles  and  locusts.  In  one  instance  I 
found  remains  of  scorpions  and  some  large  spiders ;  in 


AFES.  331 

another  there  were  fragments  of  bones,  apparently  of  a 
tortoise.  They  were  not  in  a  state  to  give  any  indication 
as  to  whether  the  animal  had  been  killed  and  eaten  by 
the  Hombill,  or  whether  its  remains  had  been  found  and 
swallowed  after  death.  I  once  saw  an  Abbagumba  near 
a  number  of  mule  carcases,  but  not,  so  far  as  I  could 
see,  feeding  on  them.  I  think  in  all  probability  it  was 
attracted  by  the  insects. 

During  the  day  the  Abbagumba  is  generally  seen, 
often  in  pairs  or  three  together,  on  open  plains.  At 
night  they  roost  upon  trees.  Their  flight  is  heavy  and 
somewhat  noisy ;  not  nearly  so  much  so,  however,  as  that 
of  the  great  Indian  Hombill,  Buceros  hicoimis,  L.,  which 
may  be  heard  at  an  immense  distance.  I  have  occasion- 
ally seen  B.  abyssinicus  on  trees  in  the  day,  but  it  does 
not  frequently  remain  upon  them. 

Their  voice  is  very  deep  and  sonorous.  They  have 
two  principal  notes ;  one,  only  uttered,  so  far  as  I  heard, 
on  the  ground,  is  something  like  "  htim-hilm,"  the  first  note 
higher  in  the  scale  and  longer  than  the  second.  The 
other  is  a  peculiar  booming  sound,  so  much  resembUng 
the  rather  distant  roar  of  a  Lion,  that  on  the  Anseba, 
where  both  lions  and  Ground  Hombills  are  common,  it 
was  some  time  before  I  could  distinguish  the  two.  This 
noise  was  made  both  on  the  ground  and  on  trees,  and  is 
frequently  heard  in  the  evening. 

Bucorvus  abyssinicus  is  chiefly  found  from  about 
4,000  feet  to  7,000  or  8,000,  but  it  is  occasionally 
met  with  at  lower  elevations,  and  I  once  saw  a  bird  at 
Komavli. 


332  ZOOLOGY. 

Family  UPUPIDiE. 

72.  Upupa  epops,  L. 

Riipp.  Syst.  Uebers.  No.  102. 

l/.  minor,  Lefebvre,  p.  86. 

U.  senegalensis,  (?)  Ferr.  et  GaL  No.  161. — Brehm,  Habesch,  No.  37 

Generally  distributed.  I  shot  one  specimen,  a  female, 
in  July  in  the  Anseba  valley,  and  saw  others ;  so  some 
must  remain'throughout  the  summer.  It  is  just  possible 
that  these  may  belong  to  Swainson's  U.  senegalensts,  and 
may  be  a  distinguishable  non-migratory  race ;  but  the 
differences  appear  to  me  of  no  importance. 

73.  Irrisor  erythrorhynchus  (Lath.). 

Upupa  ertfthrorhynchos,  Lath.,  Gren.  Syn.,  Supp.  pL  110. 
P  do.  variety  with  a  black  tail,  Salt,  No.  22,  App.  p.  xlvL 
Promerops  erythrorhynchus,  Riipp.  Syst  Uebers.  No.  103. — Ferr.  et 

Gal.  No.  158. 
P.  senegaUnsis,  Cuv.  apud  Blyth,  Report  on  ZooL   ColL   firom 
Somali  Country,  Joum.  Ab.  Soc.  Bengal,  1855,  p.  299. 

Iris  dark  brown,  legs  pink,  rather  dull  The  bill  is 
sometimes  red  throughout,  at  others  entirely  black  ;  more 
frequently  partly  one  colour,  partly  the  other :  it  is  con- 
siderably longer  (about  ^  in.)  in  the  male  than  in  the 
female.  These  birds  have  a  very  peculiar  and  rather  un- 
pleasant odour.  They  are  met  with  in  small  flocks,  which 
are  very  noisy,  and  hunt  for  their  food,  small  insects, 
about  the  trunks  and  branches  of  trees.  After  hunting  for 
some  time  about  one  tree,  they  fly,  often  singly,  one  after 
the  other,  to  another,  keeping  up  a  great  chattering. 

I  found  Irrisor  erythrorhynchus  most  comn^on  in  the 
Anseba  valley,  and  it  appears  chiefly  to  belong  to  the 


AVES,  333 

subtropical  and  the  wanner  portion  of  the  temperate 
region,  being  occasionally  seen  on  the  highlands  around 
Senafd,  &c.,  wherever  there  are  many  trees.  I  also  saw  it, 
and  shot  one  bird,  in  the  tropical  region  at  Ailat. 

The  sternum  has  a  much  lower  keel  than  that  of 
Upupa,  which  it  otherwise  resembles.  There  is  a  single 
moderate  notch  in  the  hinder  edge. 

By  comparison  I  have  ascertained  that  the  bird  from 
the  Somali  country  in  the  Calcutta  Museum,  named  Prom, 
senegalensis  by  Mr.  Blyth,  and  wliich  he  considered  dis- 
tinct from  the  Cape  /.  erythrorhynqhus,  is  a  yoimg 
female  with  a  short  bill  and  less  metallic  gloss  on  the 
feathers.  I  obtained  precisely  similar  specimens  myself. 
In  some  specimens  from  South  Africa  I  find  the  white 
spots  on  the  primaries  C[uite  as  well  developed  as  in  those 
from  Abyssinia.  The  colour  of  the  bill  varies — red,  black, 
or  the  two  colours  mixed  in  every  conceivable  proportion. 

The  following  are  dimensions  of  five  specimens  : — 


Wing. 

T.n      Whole  length. 
**"-           about 

Baifh>m 
nostril. 

in. 

in.                in. 

in. 

1. 

Male  from  Seni^iS  .    , 

.     .     6-9 

8-8           16-6 

1-8 

2. 

„        Anseba . 

.    .     6-6 

8-5           16 

1-9 

3. 

Female  from  Anseba  . 

.     6-4 

8-6           1475 

1-3 

4 

w                    w 

.     .     5-45 

8-2           14-8 

1-26 

5. 

„            Ailat  .    . 

.     5-3 

imperfect       — 

1-25 

The  last  is  immature  and  in  dusky  plumage,  almost  with- 
out iridescent  colour. 

I  presume  Salt  refers  to  this  biid,  and  that  the  black 
tail  attributed  to  it  is  a  mistake  for  black  beak.  Of 
course  he  is  in  error  in  stating  that  they  feed  on  the 
fruits  of  Ficus :  it  is  the  insects  in  the  fruits. 


Tail 

Whole  length 

BfllfromiKMtra 

ABli* 

about 

to  tip. 

in. 

in. 

ia. 

6-8 

10 

1 

334  ZOOLOGY, 

74.  I.  (Rhinapomastes)  aterrimus  (Steph.). 

Fromerops  cUerrimu$y  Steph.  Shaw's  Gen.  ZooL,  Atos,  xir.  257. 
P.  pusiUus,  Swains.  B.  West  Afr.  iL  p.  120. 

P.  eyanomdas,  Riipp.    Syst    Uebers.    No.   104. — Feir.   et  QsL 
No.  159. 

Iris  brown,  bill  black,  gape  yellow,  legs  black. 

A  silent  bird,  only  seen  solitary,  and  apparently  not 
associating  in  flocks  like  /.  erythrorhynchus.  It  was 
met  with  rarely  in  the  Anseba  valley  amongst  thick 
bushes  and  trees.     Dimensions : 

Wing. 

in. 
Male 4-3 

Female 38  6-3  875  0-85 

The  latter    is   in   duller  plumage,  and  is  perhaps  im- 
mature. 

Family  CYPSELID.E. 

75.  Cypselus  sequatorialis,  MiilL 

Miiller,  Naum.  1851,  ir.  p.  27. — Id.  Desc.  de  noav.  Oib.  d'Afr. 

tab.  viL— Sclater,  P.  Z.  a  1865,  p.  69a 
0.  Buppdlif  Heugl.  Syst  Uebers.  No.  106  (withoat  descr.) ;  Om. 

N.  0.  Afr.  p.  141,  No.  102. 
C.  meZ6a,  Lefebvre,  p.  77. — HeagL  Cm.  N.  0.  Afr.  p.  140. 

This  bird  was  not  rare  around  Senafd  in  February  and 
March,  and  I  am  inclined  to  believe,  from  having  seen 
the  birds  late  in  the  evening  and  early  in  the  morning 
flying  round  the  cliffs  of  the  trachyte  hills  (known  in 
the  army  as  Senafd  rocks)  west  of  the  camp,  that  they 
roosted  there.  In  May  they  had  disappeared ;  at  least 
I  could  see  none.  Their  flight  is  very  similar  to  that  of 
C.  melba,  and  I  think  equally  swift.  I  only  shot  two 
specimens.     The  measurements  of  one  of  thesBi  a  female. 


AFES.  335 

are  : — ^Whole  length  8 '5  in. ;  wing  7'6  ;  tail,  outer  feather 
3*4,  middle  feathers  3  ;  bill,  point  to  gape  nearly  1. 

Von  Heuglin's  attempt  to  substitute  his  name  C. 
Riippellij  published  without  any  description  in  his  "  Syst- 
Uebers."  in  the  Sitzungsbenchte  der  mathem.-natUrw. 
Classe  der  Kais.  Akad.  der  Wissenschaften  (Vienna) 
for  1856,  and  only  described  in  Cabanis'  "Journal 
fiir  Omithologie"  for  1861,  for  Von  Miiller's  name 
C.  (Equator ialisy  published  with  a  description  in  the 
Naumannia  for  1851,  and  figured  in  Von  Muller's  "  De- 
scription des  nouveaux  Oiseaux  d'Afrique,"  published 
in  1853,  deserves  especial  notice  and  reprobation.  The 
excuse  given  in  Von  Heuglin's  note,  "  Omith.  N.  0. 
Afr."  p.  142,  that  the  name  C.  cequatorialis  cannot  be 
retained  because  the  species  has  not  yet  been  observed 
south  of  1 4"*  north  lat.,  is  scarcely  valid. 

76.  C.  apus,  L. 

Riipp.  Syst.  Uebers.  No.  68. — Ferr.  et  GaL  No.  163.— HeugL  Om. 
N.  0.  Afr.  No.  loa 

This  bird  was  not  noticed  on  the  highlands.  In  the 
Anseba  valley  many  appeared  at  the  end  of  July,  none 
being  seen  before  the  25  th.  Some  of  the  specimens  shot 
had  pale  edgings  to  the  feathers,  and  were  evidently 
young  birds  ;  others  were  of  uniform  colour. 

77.  C.  afRnis,  Gray  and  Haidwlcke. 

111.  Ind.  Zool.  pi.  35,  fig.  2.— Sclater,  P.  Z.  S.  1865,  p.  604. 

C,  abysHnicus,  Streubel,  Isis,  1848,  p.  354. 

C.  galilcBen»i3y  AntinorL 

a  affinis,  Heugl.  Orn.  N.  0.  Afr.  No.  146. 

Only  met  with  at  the  base  of  the  hills  and  for  a  short 
distance  within  the  pass.     In  May  many  were  breeding 


336  ZOOLOGV, 


under  rocks  in  tbe  Suru  pass.  Brehm  states  that  this 
bird  breeds  on  palms.  Of  course  he  has  confounded  its 
nidification  with  that  of  another  group  of  Swifts,  to 
which  belong  C.  batdssiensis,  Gray,  of  India,  and  C. 
parvus,  Licht  of  Africa. 


Family  CAPKIMULGIDiE. 
78.  Caprimulgus  nubicus,  Liclit. 

Dubl.  Cat.  p.  59. 

C.  infuseaiiiSj  Riipp.  Atlas,  t.  6,  p.  9. — Syst.  Uebers.  No.  61. 

C  tamariciSf  Tristnini,  Ibis,  1866,  p.  75,  pi.  ii. 

0.  nubiciut,  Heugl.  Om.  N.  0.  Afr.  No.  92. 

I  obtained:  a  single  specimen  of  a  Nightjar  near  Zulla, 
which  may  perhaps  belong  to  this  species,  as  it  agrees 
in  measurements  and  in  most  of  the  characters  with 
RuppeU's  description.  It  diflfers  in  colour  from  the 
plate,  being  much  greyer,  but  in  the  case  of  other 
species  figured  in  Ruppell's  Atlas  the  coloration  is  not 
always  quite  accurate.  Riippell  only  mentions  a  white 
band  on  the  three  outer  primaries.  In  the  bird  from 
Zulla  the  four  first  primaries  are  banded,  the  band  being 
imperfect  on  the  first.  The  measurements  are :  Wing 
5'6  in.,  tail  4,  tarsus  0*8,  whole  leogth  about  8*75. 

I  think  there  can  scarcely  be  a  doubt  but  that  my 
bird  is  also  identical  with  C.  tamaricis,  Tristram.  That 
ornithologist  has  doubtless  been  misled  by  Riippell  s 
very  poorly  executed  plate.  It  also  agrees  with  Lich- 
tenstein's  description. 

There  are  two  white  spots  on  the  throat. 


AFES.  337 

79.  C.  inomatus,  HeagL 

Heagl.  Om.  N.  0.  Afr.  No.  94,  p.  129. 

Common  at  Ain,  Mohabar,  and  Kokai  in  the  Lebka 
valley ;  I  also  shot  one  specimen  at  Komayli.  At 
Kokai  many  of  these  Goatsuckers,  together  with  Kites, 
Rollers,  and  other  insectivorous  birds,  were  attracted  by 
a  swarm  of  winged  termites  issuing  horn  their  nest 
Measurements : — 


Wing. 

Tail. 

Tkrmu. 

in. 

in. 

in. 

6 

47 

0-6 

Male 

Female 6  4*6  07 

80.  C.  tristig^a,  Eiipp. 

Syst    Uebers.  p.    14,  t  3.--Ferr.   et    GaL   No.  161. — ^Brdim, 
Habeach,  No.  21.— HeugL  Om.  N.  0.  Afr.  No.  91. 

A  wing  obtained  near  Antalo  was  given  to  me  by 
Lieut  St.  John. 

Besides  these  Caprimulgi  a  bright  rufous  species  was 
seen  two  or  three  times  on  the  Anseba  in  thick  wood, 
but  no  specimen  was  shot.  This  may  have  been  C. 
isabellinuSy  Temm. 

Sub-Obdbr  PASSERES. 
Family  LANIADiE. 
81  •  Lanius  fallax,  Finsch. 
Finsch,  Trans.  Zool.  Soc 

This  race  is  distinguished  by  Dr.  Finsch  from  L. 
lahtoray  Sykes,  on  account  of  its  greyish  breast  aud 
abdomen.  The  younger  birds  are  dull  earthy  grey 
above,  the  adults  pure  cinereous  as  in  i.  laJitoray  and 
Dr.  Finsch  informs  me  that  he  had  only  seen  the  former. 
Lord  Walden,  who  has  kindly  examined  my  specimens, 

z 


338  ZOOLOGY. 

and  compared  them  with  his  fine  series  of  L.  lahtora^ 
tells  me  that  the  only  difference  he  can  discover  is  that 
in  the  fallax  race  the  secondaries  are  white-tipped  and 
have  the  basal  half  of  the  inner  web  white,  as  in  i.  eaxu- 
hitor ;  whilst  in  L.  lahtora  of  India,  the  whole  inner 
web  of  the  secondaries  is  white.  Further  examination  of 
specimens  from  intermediate  localities  will  be  necessary. 
Lanius  fallax  was  abundant  on  the  shores  of  Annesley 
Bay  in  January  and  February.  It  is  evidently  migra- 
tory there,  for  it  had  disappeared  in  May.  I  again 
obtained  specimens  near  Massowa  in  August. 

82.  L.  humeralis,  Stanley. 

Stanley,  in  Salt's  Voyage,  App.  p.  li. 
L.  fiscuSf  Cabanis,  Mus.  Hein.  L  p.  74. 
Abyssinian  Shrike,  Latham,  G«n.  Hist.  Birds,  iL  p.  33. 
Laniaritu  eoUairis,  Riipp.  Syst.  Uebers.  No.  230. — Ferr.  et  QaL 
No.  62. 

Iris  dark  brown,  pupil  large. 

Very  common  throughout  the  highlands,  and  seen  as 
low  down  as  the  Anseba  valley,  where,  however,  it  is  rare. 
It  is  a  true  Shrike  in  all  its  habits,  and  has  none  of  the 
skulking,  hiding  manner  of  the  bush  Shrikes.  Ruppell 
was  certainly  in  error  in  calling  it  Laniarius.  It  is 
usually  seen  sitting  on  the  tops  of  bushes,  and  then 
pouncing  down  upon  insects. 

It  breeds  in  the  rainy  season.  I  shot  a  female  con- 
taining a  completely  developed  egg  on  July  1 8th.  The 
egg  was  of  a  bluish  green  colour. 

The  most  conspicuous  distinction  between  this  race 
and  the  Cape  L.  collaris,  L.,  is  that  the  Abyssinian  bird 
has  the  breast  pure  white,  whilst  in  the  southern  form 


AVBS.  339 

it  is  greyisL  Cabanis  also  states  that  the  northern  fonn 
has  narrower  tail  feathers,  and  the  dimensions  are  rather 
smaller.     I  give  those  of  a  pair  collected  by  myself : — 


Male,  Senaf(6     .    . 

Wing, 
in. 

.    .    37 

Tail 
in. 

4-6 

Tartuft. 
in. 

0-96 

Bill  from  front 
in. 

0-68 

Female,  Adigrat 

.     .    37 

4*85 

0-98 

0-66 

I  am  somewhat  surprised  that  all  previous  writers 
should  have  overlooked  Lord  Stanley's  description  of 
this  bird.  His  measurements  of  the  bill  are  not  quite  ac- 
curate, but  otherwise  the  description  is  excellent.  I  was 
for  some  time  puzzled  by  one  character  mentioned  by 
him,  the  presence  of  a  chestnut  spot  on  the  sides  above 
the  thighs ;  but  I  find  it  in  one  of  the  specimens  before 
me,  a  female.  In  order  to  detect  it,  it  is  necessary  in 
general  to  turn  up  the  long  uropygial  feathers.  In  a 
male  specimen  I  cannot  detect  it,  so  it  is  certainly  not 
constant,  and  may  be  sexual.  It  is  noticed  by  Cabanis 
as  occurring  also  in  L.  collaris. 

83.  L.  isabellinus,  Hemp,  and  Ehr. 

Hemp,  and  Ehr.,  Symb.  Phys.,  Aves. — ^Walden,  Ibis,  1867,  p.  224, 
pi.  V.  f.  1. 

Common  on  the  coast  in  December,  January,  and 
February.  I  shot  it  also  near  Massowa  in  August.  In 
June  and  July  I  did  not  observe  it,  and  think  it  had 
migrated.  The  only  specimen  of  a  male  in  fiilly  adult 
plumage  agrees  very  well  with  the  figure  and  description 
in  the  "  Ibis."  A  female  has  the  head  much  less  rufous, 
only  the  trace  of  a  white  band  on  the  primaries,  and 
the  under  parts  pale  brownish,  with  a  pinkish  tinge.  A 
young  bird  wants  both  the  band  on  the  quills  and  the 

z  2 


340  ZOOLOGY, 

black  eye-streak,  and  has  crescentic  marks  on  the  breast 
as  in  Z.  cristatus,  &c.  The  tail  also  is  dusky.  In  all  there 
is  a  sub-obsolete  banding  of  the  rectrices.  Dimensions  : — 

Wing.  TaiL  TaraiiB.    Bill  from  front, 

in.  in.  in.  In. 

Male 37  3-56  09  -62 

Female 3-8  376  I'D  -63 

84.  L.  collurio,  L. 

Enneoctonus  coUuriOf  Riipp.  Syst.  Uebeis.  No.  211. 

Not  common.  The  only  specimen  I  obtained  was  shot 
on  the  coast  in  August. 

85.  L.  rufus,  Brisson. 

PL  enL  »,  1  2.— Riipp.  Syst.  Uebera.  No.  317. 

Only  seen  on  the  highlands,  where  it  is  rare. 

86.  L.  nubicus,  Licht. 

Lichtenstein,  Veiz.  cL  DoubL  p.  47. — Hemp,  and  Ehr.,  Symb.  Phys. 
L,  personaius,  Temm.   PL  CoL   25,  t  2.— Riipp.   Syst.   Ueben. 
No.  218. 

Seen  occasionally  about  Komayli  in  January  and 
February,  and  in  the  lower  Lebka  valley  in  August  It 
appeared  to  leave  the  tropical  region  in  the  intermediate 
period,  and  I  did  not  meet  with  it  in  the  highlands 

87.  Laniarius  sethiopicus  (Gm.). 

Merle  noir  et  blanc  d'Abyssinie,  BufL  Hist  Nat  des  Ois.  t.  iiL 

p.  406  (1776  edit).— Ibid,  tome  iv.  p.  123  (1777). 
Tardus   cethiopicus,  Gm.  Syst  Nat  i  part  iL  p.  824. — TAtham, 

Ind.  Gm.  i  p.  367. — Shaw,  Gen.  ZooL  voL  x.  p.  237. 
Lwnius  ferrugineuiy  var.  Stanley ;  Salt's  Journey,  App.  p.  61. 
L,  cethiopicuSf  Vieill.  Tab.  Enc.  et  Meth.  Gis.  p.  731,  part — ^Riipp. 

Neu.  Wirb.  p.  32. 
Tdophorus  cBthiopicus,  Riipp.  Syst  Uebers.  p.  60,  t  xxiiL — Gu^iin, 

Rev.  ZooL  1843,  p.  161.— Ferr.  et  Gal.  Voy.  en  Abyss.  iiL 

196.— Atlas,  pL  4. 

Iris  reddish  brown,  bill  black,  legs  bluish  grey. 


AFES.  341 

A  lively  active  bird,  and  less  skuUdng  in  its  habits  than 
most  bush  Shrikes.  The  male  has  a  most  melodious 
flute-like  whistle,  rarely  single,  generally  of  three  or 
four  notes,  which  must  be  familiar  to  every  Abyssinian 
traveller :  the  answer  of  the  female  is  a  hoarse  note, 
something  like  that  of  the  common  Shrike.  Besides  this, 
the  bird  has  another  call,  like  *ttik-ttik-ttik-tatatattik ' 
quickly  uttered. 

In  January  and  February  these  Shrikes  abounded, 
generally  in  families  of  four  or  five,  in  the  pass  leading 
to  SenafS  between  3,000  and  6,000  feet  None  were 
ever  seen  around  Zulla.  In  May  they  were  common  at 
a  much  greater  elevation,  even  at  8,000  and  9,000  feet 
near  Senafd,  where  none  were  met  with  three  months 
before.  They  were  occasionally  seen  in  Samhar  around 
Ailat^  &c.  in  July,  and  they  abounded  in  the  Anseba 
valley.  They  thus  appear  to  have  a  considerable  range  in 
altitude,  being,  however,  most  common  at  aU  periods  of 
the  year  in  the  subtropical  regions.  They  feed,  like  other 
bush  Shrikes,  on  various  small  insects  picked  oflF  the 
leaves  and  stems. 

In  July,  near  Kelamet  in  the  Lebka  valley,  I  saw  two 
couples  of  these  birds  together,  engaged  in  a  most  extra- 
ordinary dance-like  motion.  They  were  sitting  on  small 
branches,  moving  their  bodies  up  aujd  down,  as  if  cour- 
tesying  to  each  other,  the  males  meantime  constantly 
repeating  their  flute-like  notes,  to  which  the  females 
answered.  This  e^jtraordinary  proceeding  continued  for 
a  long  time. 

I  give  the  synonymy  of  this  bird  at  some  length. 


342  ZOOLOGY. 

because  the  species  is  generally  incorrectly  attributed  to 
Vieillot  or  Latham,  instead  of  to  Gmelin,  A  drawing 
of  the  bird  was  brought  from  Abyssinia  by  Bruce,  and 
on  this  drawing  BuflFon  founded  his  "  Merie  noir  et  blanc 
d'Abyssinie,"  on  the  description  of  which  both  Gmelin's 
Tnrdus  cBthiopicus  and  Vieillot's  Lanius  cethiopicus  were 
founded. 

There  can  be  no  doubt  also,  I  think,  that  this  is  the 
Lanms  ferrugineuSy  var.  of  Stanley. 

88.  L.  gambensis  (Liclit). 

Lanius  gambensis,  Licht.  Verz.  d.  Doabl.  p.  48^ 
?  L.  evhUi,  Stanley,  Salt's  Journey,  App.  p.  li 
Dryoscopus  cubla,  Riipp.  Syst.  Uebeis.  No.  226. 

Iris  pale  reddish  orange  (burnt  sienna),  bill  and  legs 
bluish  grey. 

Only  seen  on  the  Anseba.  A  bird  of  very  shy  retired 
habits,  much  more  so  than  the  last  species.  I  did  not 
hear  its  voice.     Dimensions  of  a  pair : —  * 

Wing.         TaiL       Tannt.       BflL 
in.  in.  in.  in. 

Male 3-6        31        0-96        07 

Female 34        305      0*9  07 

The  male  has  the  top  and  sides  of  the  head  and  heck 
black,  with  greenish  reflections,  tail  also  black  In  the 
female  these  parts  are  dusky  brown.  The  underparts 
are  purer  white  in  the  male. 

89.  L.  cruentus  (H.  and  Ehr.). 

Lanius  cruentu^s.  Hemp,  and  Ehr.,  Symb.  Phys.,  Ares,  t.  iii. 
Lanius  cruentatus,  Riipp.  Neu.  Wirb.  p.  31. 
Laniarius  eruentuSf  Riipp.  Syst  Uebers.  No.  227. 

Iris  pearl  grey,  bjU  dusky,  legs  greenish  homy. 


AFES,  343 

The  male  has  the  throat  and  centre  of  breast  pure  rose- 
coloured  ;  in  the  female  the  upper  part  of  the  breast  and 
sides  of  the  throat  are  black,  as  excellently  shown  in 
Hemprich  and  Ehrenberg's  figure,  but  they  reverse  the 
sexes  in  the  description.  This  bird  has  very  similar 
habits  to  the  next  species,  but  is  perhaps  more  easily 
seen,  as  it  inhabits  thinner  bushes.  It  was  not  rare 
around  Annesley  Bay,  and  was  occasionally  seen  in 
Samhar  and  Hahab,  up  to  an  elevation  of  about  3,000 
feet,  never,  however,  above  the  range  of  tropical  flora 
and  fauna.  In  December  and  January,  small  families 
were  met  with,  hopping  along  the  ground,  and  flying 
one  after  the  other  from  bush  to  bush  like  Crateropi 
or  Malacocerci ;  in  May  and  June,  all  met  with  were 
in  pairs.     Measurements  : — 

Wing.         Tail.       Tarsus, 
in.  in.  in. 

Male 375        4*8        IZ 

Female 365        4*4        1*3 

go.  L.  erythropterus  (Shaw). 

Lanius  erythropUrus,  Shaw's  Gen.  ZooL  Aves,  vol.  vii  p.  201. 
Laniarius  erytkrapterus,  Riipp.  Syst.  Uehers.  No.  229. 

Iris  peculiar,  brown  with  seven  white  specks  in  a  circle 
surrounding  the  pupil. 

A  skulking  bush-haunting  bird  and  quite  intermediate 
in  form  between  the  true  Shrikes  and  the  CrateropidcB. 
It  has  the  flight,  and  to  a  great  extent  the  habits,  of 
the  latter. 

In  January  and  February  I  found  the  Eed-winged  Bush 
Shrike  in  small  families ;  later  in  the  year  they  were  in 
pairs.     I  shot  a  young  biid  only  just  able  to  fly,  near  the 


344  ZOOLOGY. 

end  of  July.     I  never  saw  any  below  about  4,000  feet, 
and  they  are  common  on  the  highlands. 

91.  Nilaus  brubru  (Lath.). 

Lanius  hrvbrUy  Latii.  Ind.  Omith.  Supp.  p.  xx. 
Nilaus  brubru,  Riipp.  Syst.  Ueben.  No.  223. 

Iris  brown,  legs  bluish  grey,  rather  dark,  bill  dusky 
above,  paler  and  greyish  beneath. 

Usually  seen  in  trees.  It  was  not  very  common.  One 
specimen  was  shot  at  Ailat  in  Samhar,  and  two  or  three 
others  on  the  Anseba.  The  Tcet  are  very  weak  for  those 
of  a  Shrike. 

92.  Dicrurus  divaricatus  (Lichi.). 

Musicapa  diva/ricalaj  Lichi.  Verz.  d.  DoubL  p.  52. 
Edolius  lugubriSf  Hemp,  and  Ehr.,  Sjmb.  Phys.,  t.  viiL  3. 
Dicrurus  canipennis,  Swains.  B.  West.  Ah,  L  p.  254. 
D.  lugubrisy  Riipp.  Sjst.  Uebers.  No.  216. 

Iris  red. 

Chiefly  found  in  the  lower  part  of  the  subtropical  zone, 
but  occurs  also  in  the  coast  plain.  It  is  a  quicker  bird 
than  the  well-known  "King  Crow"  of  India,  but  has 
a  somewhat  similar  call,  heard  only  in  the  summer. 

Dr.  Finsch  considers  the  Abyssinian  species  identical 
with  Lichtenstein's  bird  from  Senegambia.  I  unfor- 
tunately omitted  to  compare  them  when  in  Berlin. 

93.  Tchitrea  melanogastra  (Swains.). 

Musicapa  paradisic  and  M.  mutaJUt,  var.  Salt,  N08. 34  and  35,  App. 

p.  xlvii 
Muscipeta  melanogastra^  Swains.  B.  of  W.  Afr.  ii  p.  55. — Riipp. 

Neu.  Wirb.  p.  108  ;    Syst  Uebere.  No.  211. 
Jfcf.  Ferreti,  Gu^r.  Rev.  Zool.  1843,  p.  162. 
Tchitrea  Ferretif  Ferr.  et  Gal.  Voy.  en  Abyss,  iii.  p.  212  ;  Atlas, 

pi.  8. — Antinori,  Cat.  Ucc  p.  46. — Joum.  f.  Om.  1867,  p.  96. 

The  variations  in  this  bird's  plumage  are  described  by 


AFES.  345 

Antinori  {loc.  dt).  It  is  by  no  means  clear,  however,  that 
they  are  due  to  age  alone.  I  have  seen  specimens  in  all 
stages,  precisely  as  with  the  Indian  species,  but  pure 
white  and  black  birds  are  very  rare.  I  never  saw  but 
one  in  Abyssinia,  and  that  I  was  unable  to  secure. 

The  Abyssinian  Paradise  Flycatcher  is  most  common  in 
the  subtropical  region,  but  is  also  found  in  warm  valleys 
on  the  highlands,  up  to  about  7,000  feet,  and  it  is 
occasionally  seen  almost  at  the  sea-level. 

I  find  that  Dr.  Finsch  has  united  the  Abyssinian  species 
to  that  described  by  Swainson  from  Senegal  I  had  pre- 
viously suspected  that  the  two  were  identical,  the  only 
diflference  I  could  detect  being  the  darker  or  paler  colour 
of  the  abdomen,  which,  probably  like  the  various  distri- 
bution of  the  chestnut,  black,  and  white  colours,  and 
especially  the  replacement  of  chestnut  by  white  in  the 
long  tail  feathers  insisted  on  by  Gu^rin-MeneviUe,  is  a 
distinction  due  to  age. 

94.  Platysteira  pririt  (Vieill.). 

Mtucicapa  pririt,  VieilL  Nouv.  Diet  d'Hist  Nat.  xxL  p.  486. 

Iris  golden  yellow. 

The  male  has  a  black  breast,  female  chestnut;  they 
were  constantly  seen  and  shot  in  pairs.  Dr.  Finsch  con- 
siders that  there  are  two  distinct  species,  P.  senegalensis 
and  P.  pririt,  VieilL  in  Mr.  Jesse's  collection.  In  this 
case  I  probably  have  them  also,  but  not  in  England. 
I  found  this  Flycatcher  from  just  above  the  level  of  the 
coast  plains  to  the  highland.  It  was  most  common  about 
May  en,  in  the  Senafe  pass,  and  in  the  Anseba  valley,  at 
elevations  of  from  3,000  to  5,000  feet. 


346  ZOOLOGY, 

95.  Bradyornis  chocolatina,  EUpp. 

Muscicapa  chocolatina,  Riipp.  Neu.  Wirb.  p.  107,  et  Syst  Ueben. 

p.  49,  t  20. — Ferr.  et  Gal.  Voy.  en  Abyss.  iiL  p.  211. 
M.fumigata^  Gu^r.  Rev.  Zool.  1843,  p.  161. 
Curruca  chocolatina,  Riipp.  Syst.  Uebers.  p.  37,  t.  xiv. 

Iris  a  peculiar  greyish  yeUow,  biU  bluish  grey. 

Not  very  rare  on  the  highlands,  generally  seen  on  trees. 
I  am  informed  by  Dr.  Finsch  that  he  has  examined  the 
original  specimen  of  Curruca  chocolativxiy  and  that  it 
also  belongs  to  this  species. 

Family  HIRUNDINID^ 

96.  Hirundo  melanocrissus,  Riipp. 

Riipp.  Syst.  Uebers.  p.  17,  t.  v. — HeugL  Cm.  N.  0.  Afr.  No.  118. 

Only  seen  by  me  at  low  or  moderate  elevations,  but 
found  at  greater  heights  by  Klippell  and  Heuglin.  I  do 
not  recollect  ever  seeing  this  bird  on  the  tableland. 

97.  H.  alpestris,  Pall. 

Pallas  Zoog.  Ros.  As.  I  p.  634.— Heugl  Oni.  N.  O.  Afr.  No.  117. 
jET.  rufula,  Temm.  Man.  d'  Om.  iiL  p.  298. 

1  shot  two  specimens  of  this  bird  at  Komayli,  in 
February. 

98.  H.  puella,  Temm. 

Fauna  Japonica,  p.  33. — HeugL  Oni.  N.  0.  Afr.  No.  120. 

H.  abystinica,  Gu^r.  Rev.  ZooL  1843,  p.  322.— Ferr.  et  Gal.  iii. 

p.    240,   No.  164 ;   Atlas,  pL  x.— Lefebvre,  p.   77.— Brehm, 

Habesch,  No.  25. 
Cecropis  itriokUoy  Riipp.  Syst.  Uebers.  p.  18,  t.  6. 

This  bird  appeared,  to  me  to  replace  H.  raelanocrissu>s 
in  the  temperate  region.  I  have,  however,  very  few 
notes  on  the  Swallows,  as  I  hod  not  many  opportunities 
of  shooting  specimens. 


1IRUND0   ftTHIOPlCA.W  BU 


JFES.  347 

gg.  H.  rustica,  L. 

Riipp.  Syst.  Uebera.  No.  70.— Heugl.  Om.  N.  O.  Afr.  No.  111. 

Common   everywhere.     This  bird   abounded   on  the 
shores  of  Annesley  Bay  in  the  middle  of  June. 

ICO.  H.  sethiopica,  W.  Blanf.  (PL  II.). 

Ann.  and  Mag.  Nat.  Hist,  Nov.  1869. 

Cecropis  rufifronSy  Lefeb^re,  p.  78. — HeugL  Syst  Uebers.  No.  113.^ 

Brehm,  Habesch,  No.  26. 
jET.  albigularis,  Strickland  apud  HeugL  Om.  N.  0.  Afr.  Na  153, 

p.  113  (nee  Strickland,  Cent,  to  Om.  1849,  pL  17)l 
H,  similis  H.  albigulari,  Strickl,   sed  conspicu   minor  et  torque 

pectorali  interrupta,  gutture  peetoregue  ni^€9emii^lav(Ui$, 
Fern.  Long,  tola  5*25  in.,  al<B  4'3,  red.  nud.  1*5,  exL  2'3,  tarsi  0*46, 

rostr,  afr,  03,  a  rictu  0*5, poll,  et  dec.  Angl. 

This  species  is  very  similar  in  coloration  to  the  South 
African  H.  cdbigularisy  Strickland  (Jard.  ^^  Contrih.  to 
Omith.y'  1849,  p.  17 — 4  with  plate\  but  conspicuously 
smaUer,  Strickland's  dimensions  being— whole  length 
7  in.,  wing  5  in.  2  lines,  median  rectrices  2  in.,  outer 
ditto  3  in.  4  lines,  so  that  the  closed  wing  in  the 
Abysainian  bird  is  nearly  an  inch  shorter.  The  pectoral 
band  also  is  imperfect  in  all  the  specimens  I  have  been 
able  to  examine.  That  the  northern  and  southern  species 
are  distinguishable  is  certain,  and  they  are  correctly 
separated  by  Von  Heuglin  in  his  new  work,  but  it  is 
equally  clear  that  it  is  a  mistake  to  apply  Strickland's 
name  to  the  north-eastern  race,  since  Strickland's  mea- 
surements and  description  are  unquestionably  those  of 
the  southern  form,  and  he  particularly  mentions  that  one 
of  the  two  specimens  in  his  possession  came  from  South 
Africa.  He  discriminates  the  species  from  H.  rujifrons  of 
Vieillot  founded  on  Le  Vaillant's  figure  of  the  Hit&ndelle 


348  ZOOLOGY. 

d  front  roux  ("Ois.  d'Air"  pi.  245,  f.  2),  with  the  whole 
throat  and  upper  breast  black,  and  adds  that  it  is  the 
H.  ncfifrons  of  Shaw.     The  synonymy  is : — 

1.  H,  rufifrorUf  VieilL   Nouv.  Diet  d'Hist.  Nat  xiv.  p.  621. — Enc. 

Meth.  p.  524. — Hirondelle  h  front  roux,  Le  Vaillant  Ois.  d'Afr.  pL 
245,  fig.  2. — Stephens,  Shaw's  Zoology,  Ayes,  x.  p.  91. — ^Bp.  Coosp. 
p.  318.    Sanded  om  Le  VailL  Ois.  d'Afr.  p.  51. 
(I  cannot  help  suspecting  that  Le  VaiUant's  plate  is  foonded  on  a 

bad  specimen  of  H,  rustiea  from  Senegal,  perhaps  restored  or 

manu&ctured.) 

2.  H.  aXbiguUvriSy  Strickland  Q^oe.  eit.), — H,  rufifrons,  VieilL  apud  Lesson, 

Traits  d'Omith.  p.  268,  nee  VieilL — H,  albigula,  Bonp.  Consp.  p.  338. 

3.  H,  cdhiopiea,  sp.  nov.  ut  supra. 

(If,  as  appears  highly  probable,  VieUlot's  name  most  be  given  up  as 
being  applicable  to  a  numufEM^ured  or  imaginary  bird,  No.  2  will 
stand  as  H,  rufifronSy  Less.  Strickland's  name  is  to  be  preferred, 
to  avoid  ambiguity.) 

loi.  H.  ruficeps,  Licht 

Verz.  d.  DoubL  des  Zool.  Mas.  Berl.  p.  58. — Ferr.  et  6aL  iiL 

No.  225. 
Cecropis  filieaudalay  Riipp.  Syst  Uebers.  No.  75. 
Ubromitus  filifera,  Brehm,  Habesoh,  No.  28. 
H.jUiferay  HeugL  Om.  N,  0.  Afr.  No.  115. 

I  only  saw  this  Swallow  once,  when  I  shot  a  pair,  a 
male  and  female,  sitting  together  on  a  spray  overhanging 
a  small  stream  near  Agula,  about  half-way  between 
Adigrat  and  Antalo.  The  outer  tail  feathers  are  very 
much  shorter  than  is  usual  in  Indian  specimens,  and  I 
am  far  from  convinced  that  the  species  are  identicaL  K 
they  be,  of  course  the  prior  name  of  H.Jiliferay  Stephens, 
must  apply. 

In  both  the  specimens  shot  by  me,  the  chin,  throat, 
and  breast  are  pale  buff,  with  an  imperfect  dusky  pectoral 
band,  njuch  more  marked  in  tljie  fepaaie,     In  jthe  original 


AFES.  349 

description  by  Lichtenstein  this  is  noticed,  "  fascia  pec- 
torali  obsolete  femiginea."  In  the  male  the  head  is  bright 
chestnut  above ;  in  the  female,  doubtless  a  young  bird, 
dark  brown.     Measurements : — 

Middle  taU     Outer  UU 
Wing.  feathen.       feathen. 

in.  in.  in. 

Male 4-6  1-3  46 

Female 4*3  1*35  1-8 

102.  Psalidoprocne  pristoptera  (Rupp.). 

Hvrundo  pristofUrck^  Riipp.  Nen.  Wirb.  p.  105,  t.  xxxiz.  t  2. 
Chdidon  prUtopUray  Riipp.  Syst.  Uebers.  Na  81. 
Attieara  pristoptera,  Brehm,  Habesch,  No.  23. 
Psalidoprocne  pristoptera^  Sclater,  P.  Z.  S.  1864,  p.  109. — HengL 
Om.  N.  0.  Afr.  No.  148. 

I  shot  but  a  single  specimen  of  this  curious  little  bird. 
I  saw  it  occasionally  both  on  the  highlands  and  in  the 
Anseba  valley.  In  habits  and  flight  it  differs  in  no 
important  respect  from  other  Swallows,  so  far  as  I 
observed. 

103.  Chelidon  urbica  (Linn.). 

Riipp.  Syst.  Uebers.  No.  80.— HeugL  Om.  N.  0.  Afr.  No.  129. 

A  single  specimen  obtained  at  KomayU  in  February. 
It  was  hunting  over  jungle  in  company  with  Hirundx) 
rustica  and  H.  melanocrissus. 

104.  Cotyle  cincta  (Bodd.). 

Hirundo  cinctay  Bodd. — HeugL  Om.  N.  0.  Afr.  No.  125. 
jET.  torquata,  Gmelin.— Riipp.  Syst  Uebers.  No.  76. 

I  twice  found  this  Martin  abundantly,  first  on  the 
shores  of  Lake  Ashangi  in  April,  and  again  about  a 
fortnight  later  on  the  banks  of  a  stream  near  Antalo. 


350  ZOOLOGY. 

105.  C.  rupestris  (Scopoli). 

HeugL  Om.  N.  0.  Afr.  No.  122. 

A  very  common  bird  in  the  rocky  passes,  and  found 
almost  from  the  sea-level  to  8,000  feet  A  specimen 
from  Senaf^  differs  in  no  respect  from  others  brought 
from  Southern  Europe.  Singularly  enough,  a  single 
specimen  obtained  by  Mr.  Jesse  appears  to  belong  to  a 
small  variety  of  the  southern  form  C  fuligul/Xy  Licht. 
I  probably  saw  both.  Heuglin  only  gives  Cotyle  obsoleta^ 
Cab.,  a  paler  form,  as  common  in  Abyssinia.  If  I 
obtained  either  of  the  other  species,  I  have  over- 
looked them. 

106.  C.  minor,  Cab. 

Cab.  Mus.  Hein.  pt  .i  p.  49.— HeugL  Cm.  N.  0.  Afr.  Na  16a 

But  a  single  specimen  was  obtained,  which  agrees  well 
with  Cabanis's  description,  except  that  the  chin  and 
throat  are  ashy  grey  instead  of  yellowish  grey.  The 
only  difference  I  can  see  between  the  bird  brought  back 
by  myself  and  a  specimen  of  C.  palustris  are  the  smaller 
size  and  the  rather  squarer  tail  of  the  former.  I  am 
indebted  to  Mr.  K.  B.  Sharpe  for  this  identification. 

This  little  Bank  Martin  was  seen  in  countless  swarms 
on  Lake  Ashangi  in  April.  They  kept  much  in  flocks^ 
roosting  amongst  reeds  at  night. 

Dimensions,  wing  3*9  in.,  tail  1*9,  whole  length  about 
4f  in. 


AFES,  351 

Family  NECTARINEDiE. 

107.  Nectarinia  habessinica,  Hemp,  and  Ehr. 

Symb.  Phya.,  Aves,  Tat  IV. — Riipp.  Syat.  Uebers.  No.  112. 

Ferr.  et  GaL  No.  175  and  176. — Brehm,  Habesch,  No.  40.— HeogL 

Om.  N.  0.  Afr.  p.  229. 
Ginnyris  gularU,  Riipp.  Neu.  Wirb.  p.  88,  t.  xxxL  1  2  (the  young 

bird). — Syst.  Uebers.  No.  HI. 

Very  common  near  the  coast,  and,  up  to  about  4,000 
feet  above  the  sea,  in  the  passes  leading  to  the  highlanda 
In  January  and  February  many  birds  were  in  the 
plumage  described  by  KUppell  as  N.  gularis.  Others, 
however,  were  in  the  full  plumage,  and  it  is  not  quite 
clear  whether  the  gularis  plumage  is  assumed  by  all 
males  after  the  breeding  season  or  whether  it  is  only  the 
livery  of  the  first  year.  I  am  strongly  inclined  to  the 
latter  opinion.  The  nest,  figured  by  Ehrenberg,  and 
which  I  also  found,  is  very  similar  to  that  of  Nectarinia 

a  a 

asiatica. 

That  N.  gularis  is  identical  with  N.  habessinica  cajmot, 
I  think,  admit  of  a  doubt.  I  only  saw  the  former  in 
places  where  the  latter  also  occurred,  I  shot  two  or  three 
specimens  when  trying  to  obtain  females  (always  more 
difficult  to  see  and  procure  than  the  males),  and  one 
specimen  which  I  shot  was  moulting  and  in  intermediate 
plumage. 

108.  N.  affinis,  Riipp. 

Riipp.  Neu.  Wirb.  p.  87,  t.  xxxi.  f.  1 ;  et  Syst.  Uebers.  No.  110.— 
Ferr.  et  GaL  No.  177. — Lefebvre,  p.  87. — Brehm,  Habesch, 
No.  41.-  Heugl.  Om.  N.  0.  Afr.  p.  232. 

This  replaces  N.  habessinica  above  3,500  to  4,000  feet 
and   extends  upwards  into   the   temperate  region.     It 


352  ZOOLOQr. 

breeds  about  May,  and  I  saw  young  birds  with  the 
parents  in  July.  The  extent  of  blue  on  the  neck 
appears  to  vary  slightly. 

109.  N.Jardinii,  Yerreaux. 

Hartlaub,  Om.  West  Afr.  No.  133,  p.  47.— HeugL  Om.  N.  O.  Afr. 
p.  227. 

I  am  indebted  for  a  specimen  of  this  Sun-Bird  to 
Captain  Sturt.  I  did  not  myself  meet  with  it  The 
specimen  given  to  me  was  shot  at  an  elevation  of  be- 
tween 5,000  and  6,000  feet  below  Senafe,  as  was  another 
obtained  by  Mr.  Jesse. 

Dimensions  of  a  male  :  Wing  2'6  in.,  tail  1*85,  tarsus 
0*65,  bill  from  front  0*7,  whole  length  about  475  in. 

no.  N.  cruentata,  Eiipp. 

Riipp.  Syst.  Uebers.  p.  26,  t.  ix. — Ferr.  et  GaL  No.  179. — Brehm, 
Habesch,  No.  43.— HeugL  Cm.  N.  O.  Afr.  p.  228. 

Not  a  common  bird  anywhere,  and  only  seen  in  the 
upper  part  of  the  subtropical  and  in  the  temperate 
region.  It  was  occasionally  met  with  about  Senafe  and 
in  the  upper  part  of  the  pass.  I  again  met  with  it  in 
the  Anseba  valley,  keeping,  however,  to  the  hills  around, 
and  Mr.  Jesse  shot  two  or  three  specimens.  It  is  an 
exquisitely  coloured  bird  when  in  full  plumage;  more 
beautiful  perhaps  than  its  more  gaudily  attired  alliea 

III.  N.  tacazze,  Stanley. 

Certhia  Uicaassey  Stanley ;  Salt's  Journey,  Appendix,  p.  58. 
Cinnyris  Uicaaze^  Riipp.  Neu.  Wirb.  p.  89,  t.  xxxL  f.  3. — Ferr.  et 

GaL  No.  177.— Lefebvre,  p.  88. 
Neetarinia  iacaxxAy  Riipp.  Syst.  Uebers.  No.  108. 
Nectarina  taccaiena,  HeugL  Om.  N.  0.  Afr.  p.  222. 

This  fine  Neetarinia  ranges  to  a  much  higher  elevation 


AFES.  353 

than  the  other  species.  It  was  common  about  Senafe 
and  Adigrat,  and  was  still  abundant  at  10,500  feet  on 
the  Wandaj  pass.  The  non-breeding  plumage  is  dull ; 
it  is  only  in  the  breeding  season  that  the  males  acquire 
their  rich  purple  colour. 

In  May  they  were  apparently  breeding  about  Senafe. 
It  was  very  difficult  to  find  the  females,  which  were 
probably  occupied  in  incubation.  A  collector  of  mine 
shot  seven  or  eight  males,  and  on  my  sending  him  out 
again  for  females  he  could  only  obtain  one. 

112.  N.  metallica,  Licht. 

Vera,  der  Doubl.  p.  16. — Riipp.  Atlas,  p.  10,t.vii. — Hemp,  and  £hr. 
Symb.  Phys.  t.  L — Riipp.  Syst  Uebers.  No.  109. — Ferr.  et 
GaL  No.  174.— Lefebvre,  p.  89.--Brelim,  Habesch,  No.  39.— 
HeugL  Om.  N.  0.  Afr.  p.  222. 

In  December,  January,  and  February,  N.  habessinica 
was  the  only  Sun-Bird  seen  near  the  coast.  The  present 
species  first  appeared  in  March.  It  was  only  met  with 
in  the  tropical  region,  but  about  Komayli  it  was 
common  in  pairs  in  June,  and  apparently  commencing 
to  breed.  I  saw  one  nest  of  cocoons,  tree-cotton,  and 
fine  grass.  It  was  suspended  from  an  acacia,  and  had 
the  usual  form,  with  an  entrance  fix)m  the  side;  there 
were  no  eggs  in  it  in  the  commencement  of  June.  After 
the  breeding  season  this  bird  probably  loses  its  long 
tail-feathers,  as  they  were  wanting  in  a  specimen  I  shot 
in  the  beginning  of  August.  I  also  met  with  N.  metal- 
lica  in  Samhar,  the  Lebka  valley,  and  very  rarely  on 
the  Anseba. 

A   A 


354  zooLOor. 

113.  N.  pulchella,  VieilL 

Riipp.  Syst.  Uebers.  No.  107.— Ferr.  et  GkiL  No.  173.— HeugL  Om. 
N.  O.  Afr.  p.  223. 

Common  on  the  Anseba,  especially  in  the  thickets  near 
the  river,  usually  upon  high  acacia  trees,  but  not  seen 
anywhere  else.  Females  were  very  rarely  noticed.  This 
bird  has  a  fine  song. 

Specimens  of  N,  pulchella  without  the  elongated  tail- 
feathers  are  often  mistaken  for  N.  liahessinica.  They 
may  be  distinguished  by  being  bronzed  green  and  not 
purple  on  the  crown  of  the  head,  and  by  the  shorter  bill 


Family  MELIPHAGLDiE. 

114.  Zosterops  poliogastra,  Heugl. 

Von  Heuglin,  Ibis,  1861,  p;  367,  pL  iiiL  ;  et  Orn.  N.  0.  Afr.  p.  412. 
Hartlaub,  Joum.  of  Ornith.  1865,  p.  9. 

I  obtained  a  solitary  female  specimen  of  a  Zosterops 
at  Dongolo,  which  I  can  only  refer  to  this  species.  It 
appears  to  differ  in  having  the  forehead  bright  yellow ;  in 
the  black  of  the  lores  extending  slightly  under  the  eye ; 
the  yellow  supercilium,  which  in  the  figure  in  the  "  Ibis  *' 
is  represented  as  much  produced  behind,  scarcely  dis- 
tinguishable beyond  the  middle  of  the  eye;  the  upper 
tail  coverts  are  not  yeUow  but  green,  the  same  colour  as 
the  back ;  and  the  general  colouring  cannot  be  correctly 
termed  **  virescenti-Jlava"  It  is  green  with  an  olivaceous 
tinge.  The  dimensions  too  are  identical.  Length  5  in., 
bill  from  front  0'4,  wing  2*5,  tail  1*8,  tarsus  0*7. 
If  new,  this  may  be  called  Z.  Jlamgula 


1      EREMOMELA    GniSEO  -  F  LAVA,  Hcu^! . 
Z      FHTLIOSCOPUS    ABYSSI  NICUS.  W.  Bis 


AFES.  355 

115.  Z.  abyssinica,  Gu^r. 

Gu^iin,  Rev.  Zool.  1843,  p.  162. — Ferr.  et  GaL  Voy.   en  Abyss. 

part  liL  p.  209 ;  Atlas,  pL  ix.  £  2. 
?*Z.  rnada^cLscarientis,  Biipp.  Syst.  Uebeis.  No.  150. 

I  shot  two  specimens  of  this  biid  in  the  Komayli  pass 
at  Mayen,  3,500  feet  aboVe  the  sea.  Dimensions  of  a 
female:— Wing  2*25  in.,  tail  17>  tarsus  0*6,  bill  at 
front  0*4  in. 

ii6.  Eremomela  griseoflava,  HeugL  (PL  III.  fig.  1.) 

Yon  Heuglin,  Journ.  £.  Orn.  1862,  x.  p.  40,  et  Cm.  N.  0.  Air.  p. 
284. 

I  only  met  with  this  lively  and  active  little  bii-d  in 
the  Anseba  vaDey  and  its  neighboui'hood,  where  it  was 
fii"st  found  by  Von  Heuglin.  It  has  much  the  same 
habits  as  Zosterops.  It  appeared  to  be  in  parties  of 
three  or  four  constantly  moving  amongst  the  bushes. 
I  preserved  a  pair,  of  which  the  following  arc  the 
dimensions  :— 


Wing, 
in. 

Tail 
in. 

Tarsus, 
in. 

Bill  fh>m  front, 
in. 

2*-0 

11 

07 

0-36 

2-0 

105 

0-68 

0-36 

Male ; 

Female 2*0 

In  coloration  this  bird  scarcely  differs  from  E.  Jlavir 
ventrisy  Sund.  It  is,  however,  smaller  in  all  its  measure- 
ments. The  position  of  Eremomela  in  the  series  is 
somewhat  dubious.  I  cannot  see  that  it  has  any  aflSnity 
with  Drymceca,  but  it  certainly  approaches  Phylloscojms 
and  Oligura.  It  appears  to  me,  however,  to  be  most 
closely  allied  to  Zosterops^  especially  to  the  grey-coloured 
form,  Malacirops  of  Bonaparte. 

•      A  A  2 


356  zooLoar. 


Family  PABIDiE. 
ziy.  Parus  leucopterus,  SwainA. 

Birds  of  W.  Afr.  il  p.  42. 

Tarus  leueonuUUf  Riipp.  Neo.  Wiib.  p.  100,  t.  xxzyu.  1  2 ;  et  Syst. 
Uebers.  No.  170.— Heugl  Om.  N.  0.  Afr.  p.  407. 

This  species  I  only  saw  in  the  Anseba  valley,  but 
Captain  Sturt  gave  me  a  specimen  shot  near  Senafd. 
Measurements : — Wing  32  to  3*25  in.,  tail  2*5  to  2*6, 
tarsus  0*8,  beak  0*5. 

Euppell  in  his  description  ("  Neu. Wirb.'*  p.  100,  note)  re- 
ferred to  Swainson's  prior  description  of  this  species,  but 
proposed  a  new  name, because  Swainsonhad  confounded  the 
Cape  species,  P.  nigery  with  it  There  can  be  no  question 
that  both  Swainson  and  KUppell  described  and  figured 
the  same  species,  for  which  Swainson's  name  is  thus 
proved  by  EUppell's  own  remark  to  be  the  oldest 
Swainson's  title-page  is  undated  (the  date  was  1837), 
and  Kiippell's  "Neue  Wirbelthiere "  was  published  at 
intervals  between  1835  and  1840. 

* 

ziS.  P.  leuconotus,  Gu^. 

Gu^r.  Rev.  ZooL  1843,  p.  162.— Ferr.  et  GaL  Voyage  en  Abyss. 

iil   p.   222;   Atlas,  pL   ix.   I  1.— Heugl.  Oin.  N.  O.  Afr. 

p.  408. 
P.  dorsatus^  Riipp.  Syst.  Uebers.  No.  253,  p.  42^  t.  xviiL 

Occasionally  seen  about  Senafd  and  Adigrat,  but  not 
very  common. 

Measurements  of  four  specimens  : — Wing  3'1  to  3*2  in., 
taii  2'55  in. 


At'ES.  357 

Family  TtJBDIDiE. 

2x9.  Turdus  simensis,  Riipp. 

T,  muiicus^  high^'ooloared  Tarietyi  Salt^  Ko.  40,  App.  p.  68. 

T,  nmentis,  Biipp.  Neib  Wirb.  p.  81,  t  xiiz«  f.  1. — Syst.  Ueben. 

No.  19a 
T,  iemimsii,  HeogL  Om.  N.  0.  Afr.  p.  380. 

Common  throughout  the  highlands,  but  not  observed 
below  5,000  or  6,000  feet  elevation.  I  did  not  meet 
with  this  Thrush  in  the  Anseba  valley.  In  the  stomach 
of  specimens  killed  in  February,  I  found  various  insects, 
including  ants.  All  the  specimens  collected  are  much 
less  rufous  beneath  than  in  Ktippell's  figure. 

In  a  male  the  wing  measured  5*2  in.,  tail  2*9,  tarsus 
1*4  ;  in  a  female,  wing  4'3  in.,  tail  2'7|  tarsus  1*3  in. 

zao.  T,  olivacinus,  Bp. 

Bonaparte,  Conap.  Gen.  Ar.  p.  fi73.'^Betigl.  Om.  K.  0.  Afr.  p.  882. 

f  T.  (ibysnnictUf  Gm.  S^st.  Nat  L  part  2,  p.  824. 

Menda  olivoMa^  Riipp.  Syst.  Uebers.  No.  194. 

Turdu$  olvv€Uieu$,  Ferr.  et  GkL  No.  54t — Lefebvie,  p.  168. 

I  obtained  specimens  of  this  Thrush  at  Senaf^,  Adigrat, 
and  Lake  Ashangi.  At  the  latter  locality  it  was  not 
ran.  It  was  nof  .en  away  ftom  the  hig^d.  and  in 
the  Anseba  valley  appeared  to  be  replaced  by  T.  pelioSy 
Bp.,  of  which  Mr.  Jesse  obtained  specimens.  A  male 
measures  : — ^Wing  4*7  in.,  tail  4*1,  tarsus  1*3  ;  a  female, 
wing  4*4  in.,  tail,  3*75,  tarsus  1*8  in. 

X2I.  Petrocincla  cyanea,  L* 

Kiipp.  Syst.  Uebers.  No.  187. — Heugl.  Joum.  f.  Omith.  1869,  p.  145. 
Monticola  eyana,  Heugl.  Orn.  N.  0.  Afr.  p.  371. 

Occasionally  met  with  on  the  highlands. 


358  ZOOLOGY, 

122.  P.  saxatilis,  L. 

Riipp.  Syst  Uebers.  No.  188. — Heugl.  Joum.  f.  Om.  p.  146. 
Monticola  saxcUiliSf  Heugl.  Om.  N.  0.  Afr.  p.  370. 

Eare  on  the  highlands.  I  am  indebted  to  Captain 
Newport  for  a  specimen  shot  at  Adigrat  I  did  not 
myself  kill  any. 

223.  P.  rufocinerea,  Eiipp. 

Saxicola  rufocinerea,  Riipp.  Neu.  Wirb.  p.  76,  t.  xrvii.  t  1  and  2. 

— Syst.  Uebers.  No.  166. 
Thamnolcea  rufocinerea,   HeugL  Jpum.  f.  Om.  1869,  p.  151 ;  et 

Ora.  N.  0.  Afr.  p.  369. 

By  no  means  rare  on  the  highlands,  and  found  as  low- 
as  about  4,500  feet.  It  appears  to  be  a  permanent 
resident,  as  I  saw  it  on  the  hills  close  to  the  Anseba 
valley  in  pairs,  in  July.  The  female  is  much  greyer 
than  the  male.  It  has  very  much  the  habits  of  a  Red- 
start, and  is,  in  coloration  also,  a  connecting  link  between 
the  RiUicilUncB  and  the  true  Thrushes.  It  is  in  no  way 
a  Saxicola,  nor,  I  think,  a  Thamnokeq,  either.  In  the 
stomach  were  small  insects.     Measurements : — 

Wing.  Tail         Tarans. 

in.  in.  in. 

Male 3-4  26  1 

Female 3*26  2'4  1 

124.  Ruticilla  phsenicura,  L. 

Riipp.  Syst.  Uebers.  No.  142. 

Turdus  phcenicurus,  Salt,  No.  39,  App.  p.  xlvii. 

Not  very  common  about  Senaf^  in  February  and 
March. 


u 


D 

p 

i 


r 


J 


AVES.  359 

125.  Ruticilla  (?)  fuscicaudata,  W.  Blanford  (Plate  IV.). 

Ann.  and  Mag.  Nat  Hist  Not.  1869. 

K  supra  brunne$centirfuicay  uropygio  ip/ogii  rt^feseenU,  remigibw 
netfihutque  fu$cit,  vix  paUidiort-margxnoHsy  macula  preondari 
nigreseentiy  (Ubido-evrcumdat4ij  fnerUOy  ffula,  tMomineque  medio 
iordide  alhi$,  pectore  et  hypocKondriis  cinmucentibus,  Bos^ 
pedtbuique  fuseU, 

Long,  aim  2*95  in,,  cauda  2*2,  iani  9*9,  toitri  a  frtmU  0*45,  a  rietu 
0*7,  iota  circa  5*5,  poll,  et  dfc  Aiigl, 

Hah,  in  Abyisinid  septentrionali. 

This  is  a  somewhat  anomalous  hird.  In  the  general 
form,  the  bill,  and  non-scuteUated  tarsi  it  agrees  with 
Ruticilla,  but  the  sombre  plumage  rather  resembles  that 
of  a  Sylvia.  The  tail  also  is  more  rounded.  It  approaches 
nearest  to  R.  {Saxicola)  familiaris,  Stephens,  and  R. 
(Erythacus)  sinuata,  Schlegel.  None  of  the  three  are 
typical  RuticillcB.  But  for  the  non-scutellated  tarsi  I 
should  have  placed  the  present  form  in  Sylvia.  It  may 
be  allied  to  Cuphopterus. 

I  obtained  a  single  specimen  only  on  the  hills  between 
the  Anseba  and  Lebka  valleys,  at  an  elevation  of  about 
4,000  feet,  in  bushes. 

I  see  that,  in  his  new  work,  Von  Heuglin  identifies 
my  new  species  with  his  Saodcola  scotocerca.  I  believe 
my  name,  published  on  the  1st  of  November,  has  priority 
over  Dr.  Von  Heuglin's,  first  "  published ''  in  the  9th — 
11th  livraison  of  his  "Omithologie  Nord-Ost  Afinkas,*' 
which  did  not  reach  England  till  the  middle  of  December, 
though  it  may  have  been  issued  in  Germany  a  week  or 
two  sooner. 


360  ZOOLOGY. 

126.  Cossypha  semirufa  (RUpp.). 

Petrocincla  semirufa,  Riipp.  Neu.  Wirb.  p.  81. 

BessonomU    temirufa,    Riipp.    Syst    Ueben.    p.    44,    t.    21. — 

Lefebyre,  p.  96. 
Cossypha  rubrocapUla,  Gu^r.  Rev.  ZooL  1843,  p.  162. 
0.  semirufay  Ferr.  et  GaL  Voy.  en  Abyss,  ill  p.  202 ;  Atlas,  pL  6.— 

HeugL  Joum.  t  Cm.  1869,  p.  147. 
Bessomis  simirufa,  HeugL  Om.  N.  0.  Afr.  p.  376. 

Not  a  common  bird,  but  generally  distributed  in  the 
temperate  region,  and  met  with  at  nearly  10,000  feet  on 
the  Alaji  pass.     I  also  saw  it  on  the  Anseba. 

A  specimen  from  Guna  Guna  measures,  wing  3*3  in., 
tail  3-3  in. 

127.  Cercotrichas  erythropterus  (Gm.). 

Turdus  erythropterusy  Gm.  Syst  Nat  L  p.  835. 
Cercotrichas  erythropterus^  Riipp.  Syst.  Uebera.  No.  196. — HengL 
OriL  N.  0.  Afr.  p.  280. 

Common  amongst  bushes,  especially  along  the  banks  of 
stream  beds  in  Samhar  and  the  country  around  Annesley 
Bay.  Its  habits  are  precisely  those  of  the  Indian  Tham- 
nohicB ;  it  keeps  much  to  the  ground,  and  has  a  habit  of 
jerking  its  tail  completely  over  its  back,  especially  when 
stopping  after  running  or  on  settling  from  flight  It 
appeared  to  me  to  be  more  of  a  ground  bird  than  the 
Indian  Shama,  Copsychus  macrourus. 

Z28.  Thamnolsea  albiscapulata,  Riipp. 

Saxicola  albiscapulata^  Riipp.  Neu.  Wirb.  p.  74,  t  26,  f.  1. 
Thamnobia  aUnscapulata,  Riipp.  Syst  Uebera.  No.  168. 
Thamnolaa,  dlhiscapuhiay  HeugL  Journ.  f.  OmitL  1869,  p.  149 ;  et 
Ora.  N.  0.  Afr.  p.  367. 

Usually  found  near  water,  especially  about  the  rocky 
banks  of  streams.     It  was  seen  from  the  sea-level  to  the 


AF£S.  36 1 

higUlands,  and  was  common  in  the  pass  wherever  water 
was  to  be  found,  as  in  the  Sooroo  defile.     It  feeds  on 
insects :  on  one  occasion  I  found  ants  in  the  stomach. 
Measurements : — 

Wins.        lUL        Tanas, 
in.  in.  in. 

Male.  4*65        3*66        IS 

Female 4*4  3*5  11 

129.  Thamnobia  melsena  (Rtipp.). 

Saxicola  melcena,  Riipp.  Neu.  Wirb.  p.  77,  t.  21,  f.  2  ;  Syst  Ueben. 

No.  154.— Heugl.  Joum.  f.  Om.  1869,  p.  153. 
Penthokea  mdanui^  HeugL  Orn.  N.  0.  Afir.  p.  364. 

Pretty  common  on  the  highlands,  keeping  much  to 
rocky  places  amongst  bushes.  It  has  much  the  habits 
of  the  Indian  Thamnobice,  and  jerks  its  tail  like  them. 
In  the  stomach  I  found  only  insects,  principally  small 
Coleoptera. 

130.  Saxicola  cenanthe,  L. 

Riipp.  Syst  Uebers.  No.  161. — HeugL  Joom.  f.  Om.  1669,  p.  158 ; 
et  Orn.  N.  0.  Afr.  p.  347. 

Only  seen  about  Senaf^  and  Adigrat  on  the  highlands 
in  March.  It  was  not  noticed  earlier,  and  appeared  to 
be  migrating  northwards. 

131.  S.  isabellina,  EUpp. 

Riipp.  Atlas,  p.  52,  t.  34,  f.  2 ;  Syst.  Uebeis.  No.  159.— HeugL 
Joum.  f.  Om.  1869,  p.  157  ;  et  Om.  N.  0.  Air.  p.  344. 

Equally  common  on  the  highlands  and  in  the  plains 
on  the  coast  until  the  middle  of  MarcL  All  other 
species  appeared  restricted  to  one  region  or  the  other. 

There  appears  to  have  been  much  confusion  about  this 


362  ZOOLOGY. 

species,  and  I  have  seen  specimens  of  S.frenala  which 
were  distributed  by  Riippell  himself  for  it 

133.  S.  frenata,  HeogL 

Joiim.  f.  Om.  1869,  p.  168  ;  et  Om.  N.  O.  Afr.  p.  345. 

It  is  extraordinary  that  this  bird  should  have  remained 
until  so  recently  unnamed.  Specimens  of  it  are  to  be 
found  in  many  museums,  and  both  in  London  at  the 
British  Museum  and  at  Calcutta  there  are  examples 
received  from  Dr.  Rtippell  as  S.  isaheUina.  Indeed  I 
think  Dr.  Rtippell  must  have  considered  this  bird  as  the 
summer  plumage  of  that  species.  Mr.  Tristram,  however, 
informs  me  that  he  has  a  smaller  race  labelled  Saodcola 
sordida  by  Dr.  Ruppell.  The  true  S,  sordida  is  a  very 
different  bird. 

I  only  met  with  S.  frenata  on  the  very  highest 
portions  of  the  Wadela  plateau,  near  Saintora  and  Gazoo, 
at  an  elevation  of  10,500  feet  above  the  sea;  there  it 
abounded.  Von  Heuglin  states  that  he  has  seen  it  at 
a  much  lower  elevation  al^o. 

X33.  S.  deserti,  Bupp. 

Temm.  PI.  CoL  360,  f.  2.— Riipp.  Syst  Uebers.  Na  163.— HeugL 
Joum.  f.  Orn.  p.  161  ;  et  Chn.  N.  0.  Afr.  p.  362. 

Only  seen  close  to  the  coast.  Abundant  in  December 
and  January  about  Annesley  Bay,  but  it  became  scarce 
and  appeared  to  migrate  in  February.  There  were  none 
in  May,  June,  July,  or  August  near  Annesley  Bay  and 
Massowa.  This  species  can  scarcely  be  a  permanent 
resident  on  the  coast  at  least,  as  Von  Heuglin  appears 
to  think  probable. 


AVES.  365 

and  therefore  belongs  probably  to  this  race.  The  tail  is 
a  little  shorter,  very  much  more  rounded,  and  the  rec- 
trices  are  of  rather  different  shape,  being  more  pointed 
In  P.  pastor  the  central  rectrices  are  only  white  at  the 
extreme  base,  and  on  the  outermost  pair  nearly  1^  in.  at 
the  end  is  black ;  whilst  in  the  specimen  from  Annesley 
Bay  the  white  on  the  base  of  the  central  rectrices  extends 
neai'ly  to  the  end  of  the  coverts,  and  on  all  the  outer 
rectrices  only  about  f  in.  at  the  end  is  black.  The  junc- 
tion of  black  and  white  is  zigzagged,  the  black  running  to 
a  point  up  the  quill  and  the  inner  margin,  the  white  ex- 
tending further  down  on  the  outer  margin  and  the  centre 
of  the  inner  web.  The  middle  rectrices  are  0*2  in.  longer 
than  the  outer  ones.  The  bill  appears  to  be  very  slightly 
more  slender.  I  can  see  no  other  differences.  The 
measurements  are : — 

Wing.  TaiL  Tkuros.  BiH. 

in.  in.  in.  in. 

2-85  1-97  0-9  0-42 

140.  P.  semitorquata,  Hengl.  (Plate  V.). 

Joum.  f.  Om.  1869,  p.  166,  et  Om.  N.  0.  Afr.  p.  341. 

I  had  named  and  described  this  bird  at  a  meeting  of 
the  Zoological  Society  of  London  when  Von  Heuglin's 
description  appeared  in  the  "  Journal  fiir  Omithologie." 
He  is  only  acquainted  with  the  male ;  I  therefore  append 
a  description  and  figure  of  both  sexes  : — 

Mas.  niger,  macula  stcbqicadrata  collari  laterally  fascia 
alari  longitudinali  ex  remigihus  secundariis  ultimis 
cum  earum  tectricibus  compositd,  abdomine  toto  cum 
hypochondriis   uropygioque   alhis;    rostra   pedihusque 


364  ZOOLOGY. 

thick  jungle,  nor  does  it  perch  on  the  tops  of  bushes  like 
the  true  Saxicoke.  Cercomela  melanura  is  common  at 
Aden,  and  apparently  throughout  the  district  of  the 
Red  Sea. 

Z37.  Pratincola  rubetra,  L. 

Biipp.  Syst.  Uebers.  No.  160.--Het]gl  Joarn.  t  Om.  1869,  p.  167  ; 
et  Om.  N.  0.  Afr.  p.  338. 

T  only  once  met  with  this  species,  when  I  shot  a 
young  bird  at  Rairo  in  Habab,  about  the  middle  of 
August,  at  an  elevation  of  3,000  feet  above  the  sea. 

Z38«  Pi  pastor,  Strickland. 

Strickland,  Ann.  Mag«  Nat  Hist  1844,  xiiL  p.  4ia 
Sewicola  pratincola  sybUUif  Heu£^  Joum.  f.  Om.  1869,  p.  188  ;  el 
Om.  K.  O.  Afr.  p.  341. 

Not  rare  on  the  highlands.  As  mentioned  by  Von 
Heuglin,  this  bird  appears  to  breed  in  Abyssinia.  I  shot 
specimens  at  Senaf^  at  the  end  of  May.  I  did  not 
obtain  P.  rubicola.  All  my  specimens  have  a  broad 
white  uropygium,  more  or  less  tinged  with  ferruginous. 
Dimensions : — 

Wing.       TdL       Tinras.         BOt 
ixL  In.  in.  in. 

Male,  Adigrat    .    2*9        2*1        0*85        0*46  (bill  abnormally  long 

„     Senaf^      .    2*96      2*1        0'85        0*37 

Female,  SenafS .    2*8        2*03      0*86        0*43 

139.  P.  Hemprichii,  Ehr. 

SasDicola  Hemprichii,  Ehr.  Symb.  Phys. — HengL  Jonm.  f.  Omith. 

1869,  p.  168. 
Pratincola  rubicola  JSemprichiiy  HeugL  Om.  K.  0.  Afr.  p.  339. 

A  single  male  specimen  from  the  shores  of  Annesley 
Bay  differs  from  those  from  the  highlands  by  having 
much  more  of  the  basal  portion  of  the  tail-feathers  white. 


A 


AVES.  365 

and  therefore  belongs  probably  to  this  race.  The  tail  ip 
a  little  shorter,  very  much  more  rounded,  and  the  rec- 
trices  are  of  rather  different  shape,  being  more  pointed. 
In  P.  pastor  the  central  rectrices  are  only  white  at  the 
extreme  base,  and  on  the  outermost  pair  nearly  1^  in«  at 
the  end  is  black ;  whilst  in  the  specimen  from  Annesley 
Bay  the  white  on  the  base  of  the  central  rectrices  extends 
nearly  to  the  end  of  the  coverts,  and  on  all  the  outer 
rectrices  only  about  f  in.  at  the  end  is  black.  The  junc- 
tion of  black  and  white  is  zigzagged,  the  black  running  to 
a  point  up  the  quill  and  the  inner  margin,  the  white  ex- 
tending further  down  on  the  outer  margin  and  the  centre 
of  the  inner  web.  The  middle  rectrices  are  0*2  in.  longer 
than  the  outer  ones.  The  bill  appears  to  be  very  slightly 
more  slender.  I  can  see  no  other  differences.  The 
measurements  are : — 


wing, 
in. 

T&iL 
in. 

in. 

BiH. 

in. 

2-85 

1-97 

0-9 

o-4i 

140.  p.  semitorquata,  Heugl.  (Plate  V.). 

Joum.  f.  Om.  1869,  p.  166,  et  Om,  N.  0.  Ah.  p.  341. 

I  had  named  and  described  this  bird  at  a  meeting  of 
the  Zoological  Society  of  London  when  Von  Heuglin's 
description  appeared  in  the  "  Journal  fiir  Omithologie." 
He  is  only  acquainted  with  the  male ;  I  therefore  append 
a  description  and  figure  of  both  sexes  : — 

Mas.  nigeVy  macula  subquadrata  collari  lateralis  fascia 
atari  longitudinali  ex  remigihus  secundariis  ultimis 
cum  eai^m  tectricihus  compositd,  abdomine  toto  cum 
hypochondriis   uropygioque   albis;    rostro   pedihusque 


366  ZOOLOGY. 

nigris.     Long,  ales  2*65,  cavdcB  1*95,  tarsi  0*85,  rostr.  a 
fr.  0*4,  tola  circa  5  2>olh  Angl.  et  dec. 

Fomiina  supra  brunnea,  plumis  rufo-inarginatis ; 
superciliis^  mento,  gula  superiori  ahdomeneque  isabellinis, 
hoc  pallidiori,  fctscia  alari  hngitvdinali  et  uropygio 
alhis,  pectore  ferrugineO,  remigibus  cum  tectricibus 
alarum  fusco-hrunneis^  vix  pallide  niarginatis,  rectH- 
cihus  nigricantihui ;  a  mari  dimensionihus  non  discrepat. 

This  bird  is  very  near  the  Indian  P.  caprata,  L.  The 
male  is  at  once  distinguished  by  the  white  mark  at  the 
side  of  thd  nfeck,  the  female  by  being  much  more  rufous 
on  the  linder-parts.  It  is  not  a  rare  bird  in  Abyssinia, 
and  it  is  surprising  that  it  has  remained  so  long  unde- 
Scribed.  I  collected  three  pairs;  I  first  met  with  it  near 
Adigrat,  where  it  was  far  from  scarce,  and  I  again  met  with 
it  on  some  of  the  passes  south  of  Antalo,  and  at  Lake 
AshangL  It  was  never  noticed  below  about  8,000  feet 
of  elevation,  but  at  the  same  time  I  did  hdt  observe  it  in 
the  higher  plateaux,  so  that  it  appears  to  belong  to  the 
temperate  rather  than  to  the  subalpine  faiina. 

« 

141.  P.  sordida  (Kiipp.). 

Scmcola  sordida,  Riipp.  Neu.  Wirb.  p.  75,  t  xxvL  £  2.  et  Syst. 

Uebere.  No.   176.— Heugl.  Jdiirn.  t  Om.   1869,  p.  164;  et 
Om.  N.  0.  Afr.  p.  342. 

This  Bush  Chat  belongs  to  the  subalpine  fauna,  and  was 
never  seen  below  9,000  feet.  Above  that  height  it  was 
common.  It  kept  much  to  rocky  parts  of  the  hills.  It 
is  by  no  means  a  typical  Pratmcola,  the  bill  being  rather 
larger  and  the  tarsi  mnch  longer,  but  it  is  much  nearer 


AFES,  367 

to  that  genus  than  to  true  Scucicola.     The  following  are 
the  dimensions  of  a  pair : — 


Wiiig. 
in. 

TaU. 
in. 

TamuL 
in. 

Bill  from  front, 
in. 

2-66 

1-6 

0-95 

06 

276 

I'B 

11 

0-46 

Male 

Female 2  76 

142.  Grandala  leucogaster  (Gm.). 

Turdus  leucogasteTy  Gm.  Syst  Nat.  vol;  i.  pt  iL  p.  819. 
LamprotornU  leucogaster,  Riipp.  Neu.  Wirb.  p.  24  ;  et  Syst.  Uebers. 

No.  245. 
Pholidauges  leucogaMter,  Cab.   Mud.   Heim    i.   p.    198.  —  Brehm, 

Habesch,  No.  85.— Heugl.  Joum.  f.  Om.  1869j  p.  KV. 

Iris  dark  grey  within,  and  an  outer  circle  of  yfellowish 

grey- 

This  beautiful  bird  appears  to  me  fairiy  congeneric 

with    the    Himalayan    Grandala    cosUcolor,   Hodgson. 

Both  are  of  very  dubious  aflSnity,  but  their  relationship 

to   the   Starlings  is  at  least  as  questionable  as  to  the 

SaxicolincB.     Their  very  short  legs  contrast  with  both, 

whilst  the  form  of  the  wing  is  quite  different  from  that 

found  amongst  the  short-legged  Bulbuls  and  Orioles  to 

which   the  present  form  might  otherwise   be   assigned 

The  habits  are  rather  Starling-like. 

There  appears  to  have  been  some  doubt  on  the  part  of 

previous  observers  as  to  whether  the  plain  thrush-like 

plumage  is  the  livery  of  the  female  or  only  of  the  young. 

The    latter  is   stated   by   Rlippell   and   Heuglin.     The 

former  was  the  case  in  all  the  birds  shot  by  me,  and  a 

very  large  number  of  pairs  seen  by  me  in  May  consisted 

of  a  brilliantly-plumaged  male  and  a  dull  female.     Mr. 

Jesse  and  Captain  Sturt,  who  both  frequently  saw  and 

shot  the  bird,  cnme  to  precisely  the  same  conclusion,  and 


368  ZOOLOGY, 

it  is  this  diflFerence  of  plumage  in  the  two  sexes  and  the 
colouring  of  the  female  which  tend  to  unite  this  bird  so 
closely  with  Grandala  ccelicolor. 

From  December  up  to  the  end  of  February  this  bird 
was  entirely  absent  ftom  the  passes  below  Senafd.  About 
the  conmiencement  of  March  flocks  began  to  make  their 
appearance,  and  the  numbers  gradually  increased.  The 
flocks  consisted  of  males  and  females.  In  May,  when  I 
returned  from  the  highlands,  I  found  these  lovely  birds 
most  abundant,  and  almost  always  in  pairs.  They 
occurred  from  about  2,000  feet  elevation  up  to  5,000  or 
6,000  feet  I  again  met  with  them  in  the  Lebka  valley 
at  very  nearly  the  same  elevations,  one  bird  being  seen 
rather  lower  down  at  Ain ;  but  individuals  were  scarce 
below  about  2,000  feet.  On  the  Anseba  it  was  occasion- 
ally seen,  all  the  birds  noticed  being  males.  The  females 
were  probably  in  July  engaged  in  incubation.  I  could 
never  find  the  nest. 

Grandala  leucogaster  keeps  much  to  trees  and  high 
bushes,  but  is  occasionally  seen  on  the  ground.  Its  grin- 
cipal  food  appears  to  consist  of  seeds.     Dimensions : — 


Wing. 

Tail. 

TaiTOB. 

in. 

in. 

in. 

4*2 

275 

0-9 

Male 

Female 3*9        2*4  0*9 

The  male  is  perhaps  the  handsomest  of  all  the  birds 
met  with  in  Abyssinia.  The  play  of  colours  in  the  sun- 
light from  pure  azure  to  deep  crimson,  and  again  to 
copper  colour,  is  most  exquisite.  Brehm  is  incorrect  in 
saying  that  the  iris  is  brown. 


AFE&  369 

Family  PYCNONOTID^ 

Z43.  Pycnonotus  arsinoe  (Licht.). 

TwrduB  artinoSy  Licht  Yerz.  d.  Doubl.  p.  39. 
1x0$  aarnnoiy  Riipp.  Nea.  Wirb.  p.  83. 
PycTumotus  artinoe,  Riipp.  Sjst  Uebera.  Na  196. 

Habits,  song,  &c.  much  like  those  of  Indian  species. 
This  Bulbul  abounds  at  the  base  of  the  hills,  and  ranges 
to  the  highlands.    It  lives  on  fruits  and  buds. 

Family  OKIOLIDiE. 

144.  Oriolus  monacha  (Gm.). 

Turdus  monacha^  GmeL  Syst.  Nat  L  pt  il  p.  824. — Lath.  IncL 
OmiUL  L  357.— Id.  Gen.  Hist  iiL  p.  142  {Nun  Oriole). 

0.  monacha,  Salt,  No.  10,  App.  p.  xlvL 

0.  moloxUaf  Riipp.  Neu.  Wirb.  p.  29,  t.  xil  fl  1. — Lefebyre,  p.  169. 
— Riipp.  Syst.  Uebers.  No.  206. 

0.  monachtu,  HeagL  Joum.  t  Om.  1868,  p.  325. 

Iris  duU  red  ;  beak  pinkish  brown  ;  legs  bluish  grey. 
Not  a  common  bird  in  the  country  traversed,  and  I 
only  obtained  two  specimens.  It  has  a  peculiar  harsh 
double  call  note.  The  measurements  rather  exceed 
those  given  by  Rtippell.  In  a  female  shot  near  Antalo 
the  wing  is  35  in.,  tail  3*9,  tarsus  0*8,  bill  from  point 
0-8  in. 

Rtippell,  in  the  note  preceding  his  description  of 
0.  moloadta,  after  mentioning  the  descriptions  of  the 
Abyssinian  and  South  African  birds,  states  that  Latham 
named  the  latter  0.  monachtcs^  and  added  the  mohxita  of 
Buffon,  described  from  Brace's  drawing,  as  a  synonym. 
Precisely  the  reverse  is  the  case.     Latham,  in  the  "  Index 

B  B 


370  ZOOLOGY. 

Omithologicus,"  i.  p.  357,  refers  to  Buffon  alone,  and  it  is 
only  in  the  latter  work,  the  "  General  History  of  Birds/' 
that  he  adds  a  reference  to  Le  Vaillant's  figure.  Latham, 
moreover,  as  Ruppell  ought  to  have  noticed,  copies  the 
name  from  Gmelin,  whose  Turdiis  monacha  is  described 
from  Buffon's  '^moloxita,  ou  Religieuse  d'Abyssinie,"  so 
that  there  can  be  no  question  of  Gmelin's  name  applying 
to  the  species.  Von  Heuglin's  change  of  the  name  into 
O.  vnonachus  is  also  inaccurate. 

O.  monacha  is  easily  distinguished  from  the  black- 
headed  oriole  of  the  Cape,  O.  larvatus^  Licht.  by  the  paler 
colour  of  the  quills,  the  edges  of  the  secondaries  being 
pale  grey  instead  of  white.  The  bill,  too,  is  shorter  and 
stouter  in  the  northern  form. 


Family     CRATEROPIDiE,    Swainson.     (TIMALIDiE,    Bonaparte 

and  Jerdon.) 

To  this  family  I  most  unhesitatingly  refer  the  Timalince 
of  Indian  writers.  No  one  who  has  seen  both  in  their 
native  state  would,  I  think,  have  any  hesitation  on  the 
subject.  But  besides  the  African  Crateropi,  the  Indian 
Malacocerci,  Timalice,  Garrulaces,  and  their  allies,  I 
consider  that  certainly  Drymceca  and  Cisticola,  and 
probably  Prinia,  Orthotomus,  and  the  allied  forms  should 
be  classed  here,  and  not  in  the  Si/lviadw,  that  refitgium 
peccatorum  of  modern  ornithologists.  I  have  stated 
my  reasons  elsewhere,  and  need  not  repeat  them  here. 
The  Crateropi  are  very  closely  allied  to  the  Bush 
Shrikes,  Telephonus,  LaniariuSy  &c. 


AFES,  371 

145.  Crateropus  leucopygius  (Riipp.). 

1x08  leucopygius,  Riipp.  Neu.  Wirb.  p.  82,  t.  xxx.  f.  1. 

Crateropus  leucopygius,  Riipp.  Syst.  Uebers.  No.  199. 

C.  limhatus,  Rupp.  Syst.  Uebers.  p.  48,  and  No.  202,  p.  60. 

Iris  orange  yellow,  almost  scarlet;  beak  black;  legs 
olivaceous. 

Occurs  in  small  flocks  not  exceeding  eight  or 
ten  in  number,  constantly  employed  hunting  about 
on  the  ground  and  amongst  the  branches  of  trees 
and  bushes  for  their  food,  which  consists  chiefly  of 
insects.  These  birds  fly  one  after  the  other  from  tree 
to  tree,  and  every  now  and  then,  usually  when  hidden 
in  the  middle  of  a  thick  bush,  burst  out  in  a  chorus 
of  chattering,  precisely  in  the  same  manner  as  does 
Garrulax.  Indeed,  the  resemblance  in  general  appear- 
ance, flight,  habits,  and  voice  of  these  birds  to  some 
of  the  white-headed  species  of  GaiTulax  inhabiting 
the  Himalayas  and  Burma  are  most  striking,  and 
there  is  also  considerable  similarity  to  some  of  the 
3IaJacocerci. 

Crateropus  leucopygius  is  found  from  about  3,500  or 
4,000  feet  above  the  sea  to  8,000  or  9,000  feet.  It  was 
common  throucrhout  the  hiojhlands  on  the  route  traversed. 
There  can,  I  think,  be  no  question  of  C.  Umbatus  of 
Ruppell  being  merely  the  young  of  this  bird.  The  two 
certainly  occurred  together,  and  I  believe  in  the  same 
flock. 

Dimensions  : — Total  length  about  10  to  10*5  in.,  wing 
4-75,  tail  4-5  to  475,  bill  O'S,  tarsus  1*4  in. 

B  B  2 


372  ZOOLOGY. 

146.  C.  leucocephalus  (Kiipp.). 

Turdoides  Uucocq>hala,  Riipp.  Atlas,  p.  6,  t.  iv. 
Ixos  leucocephalaf  Riipp.  Neu.  Wirb.  p.  84. 
Crateropus  leucocepkalusy  Riipp.  Syst  Uebers.  No.  198. 

Iris  yellow,  with  a  slight  orange  tinge ;  bill  black ; 
legs  purplish  horn-colour. 

This  species  replaces  the  last  below  3,000  feet  It  is 
found  up  to  about  4,000  or  5,000  feet,  but  was  not  seen 
on  the  highland^  In  the  Anseba  vaUey  at  about  4,000 
feet,  both  species  were  met  with  inhabiting  the  same 
thickets.     Their  habits  are  precisely  similar. 

Dimensions : — Male,  length  about  9*25  in.,  wing  4*3, 
tail  4*75,  tarsus  1*4,  bill  from  forehead  075.  Female, 
length  about  9  in.,  wing  4 '25,  tail  4  (perhaps  not  fiilly 
grown),  tarsus  1*4,  bill  0*75. 

147.  Malacocercus  (Chatarrhoea)  acacise  (licht). 

MaXurus   acacim^  Licht.   Yeiz.    d.   Donbl.  p.  40. — Riipp,  Atlas, 

t,  xviii 
BphcMtra  aeacuBy  Riipp.  Neu.  Wirb.  p.  115. — Syst.  Uebeia.  No.  203. 
Argya  aeacicBy  HeogL  Dm.  N.  0.  Afr.  p.  387. 

It  would  be  difficult  for  any  one  to  recognise  this  bird 
from  RUppeU's  figure,  which  is  far  too  grey.  It  ought  to 
be  very  much  more  rufous.  I  only  obtained  one  speci- 
men, and  I  very  rarely  saw  the  bird  about  ZuUa.  It 
appeared,  so  far  as  I  had  opportunities  of  observing,  to 
have  precisely  the  appearance  and  habits  of  Malacocerctis 
caudatus  of  India,  which  is,  I  think,  unquestionably  con- 
generic, being  one  of  Ihe  numerous  instances  in  which 
the  forms  of  the  Indian  peninsula  are  closly  related  to 
African  species. 


AFES.  373 

The  measurements  of  the  specimen  obtained,  a  female, 
are : — ^Whole  length  about  10  in.,  wing  3*8,  tail  5*2, 
tarsus  1*25.  The  specimen  was  shot  by  a  collector,  and 
I  did  not  note  the  colours  of  the  soft  parts. 

248.  Drymceca  gracilis  (Licht). 

Malums  grcbcUis,  Riipp.  Atlas,  p.  3,  t.  ii  b. 

Drymoica  grcMiSy  Riipp.  Sjst  Uebers.   No.  117. — HeugL   Ibis, 

1869,  p.  91,  et  Om.  N.  0.  Afr.  p.  242. 
?  Bumesia  kpida,  Bljth,  Joum.  As.  So<v  Beng.  xui  p.  376^  and 

xyI  460. — Jerdon,  Birds  of  India,  iL  p.  85. 
Sylvia  graeUiSy  Licht  Yen.  d.  DoubL  p.  34. 

Iris  burnt  sienna  (orange  brown) ;  beak  black ;  legs 
flesh-colour ;  tail-feathers  10.  Very  common  amongst  the 
low  bushes  of  the  plain  country  near  Zulla. 

This  bird  has  the  usual  hiding,  skulking  habits  of  the 
genus,  and  consequently  it  was  by  no  means  easy  to  pro- 
cure specimens.  In  general,  it  is  only  see^  for  an  instant 
flying  with  a  weak  uncertain  flight  from  one  bush  to 
another,  which  it  immediately  enters,  and  is  lost  amongst 
the  roots  and  branches.  It  is  always  solitary  or  in 
pairs,  never  in  flocks.  Mr.  Jesse  found  the  nest,  with 
two  young  birds  apparently  only  hatched  a  few  days 
before,  on  Jime  12.  It  was  neatly  co^structed  chiefly 
of  grass,  with  the  entrance  at  the  side,  and  placed  in  a 
hedge  which  surroimded  the  Commissariat  enclosure. 

149.  D.  mystacea  (Hiipp.). 

Prinia  mistacea,  Riipp.  Neu.  Wirb.  p.  110, 

D.  mygta4^ea,  Riipp.  Syst.  Uebers.  p.  31,  t.  x. — Ferr.  at  GaL  No.  62. 
— Heugl.  Ibis,  1869,  p.  88  ;  et  Orn.  N.  0.  Afr.  p.  239. 

A  solitary  specimen,  a  male,  was  only  obta-ined  of  this 
species  also.  This  was  shot  on  the  highlands  at  Adabagi, 
two  marches  south  of  Adigrat. 


374  ZOOLOGY, 

Dimensions  : — Wing  2*05  in.,  tail  2*3,  tarsus  0*8,  bill 
0*45,  tail-feathers  apparently  10,  but  the  tail  is  rather 
imperfect  in  my  specimen. 

150.  D.  pulchella  (Rilpp.). 

McHunis  ptdchellas,  Riipp.  Atlas,  t.  xxxv.  fig.  a. 
Drymoica  pulchella,  Riipp.  Syst.  Uebers.  No.  118. — Ferr.  et  GaL 
No.  63.— HeugL  Ibis,  1869,  p.  92 ;  et  Om.  N.  0.  Afr.  p.  24a 

The  label  of  the  single  specimen  obtained  has  un- 
fortunately been  lost,  and  I  have  forgotten  the  exact 
locality.  I  believe,  however,  that  the  bird  was  shot  in 
the  Anseba  valley. 

This  is  by  no  means  a  typical  Drymceca,  although  it 
has  only  10  tail-feathers.     The  bill  is  peculiarly  small. 

Length  of  wing  1*85  in.,  tail  1*7,  tarsus  0*65,  bill 
0  3  in. 

151.  D.  ruBfrons  (Riipp.). 

Frinia  rufifrons,  Riipp.  Neu.  Wirb.  p.  110,  t.  xli  £  2. 

Drymoica  rufifrons,  Rupp.  Syst.  Uebers.  No.  121. — Ferr.  et  GaL 

No.  61.— Brehm,  Habesch,  No.  44.— HeugL  Ibis,  1869,  p.  87. 
Cisticola  nififrons,  HeugL  Cm.  N.  0.  Afr.  p.  245. 

Two  or  three  specimens  of  this  pretty  little  species 
were  shot,  but  only  one  adult  preserved.  Its  habits  are 
unlike  those  of  most  of  its  allies,  and  greatly  resemble 
those  of  the  little  Prinia  gracilis  of  India.  It  is  found 
in  small  flocks  amongst  low  trees  and  bushes  in  the  coast 
country  about  Annesley  Bay.  It  hides  less  in  thickets 
near  the  ground  than  most  of  its  allies ;  indeed  its  habits 
generally  are  more  those  of  Afalacocerci  or  Crateropi 
than  of  DrymceccB.  It  has  12  tail-feathers,  but  I  cannot 
agree  in  chissiug  these  bii'ds  as  Drymcecce  or  Cisticolce, 
solely  with  regard  to  the  number  of  rectrices. 


AVES.  375 

152.  Graminicola  Le  Vaillantii  (Smith). 

DrymcRca  Le  Vaillantu,  Smith,  IlL  S.  Afr.  ZooL  Aves,  t.  IxxiiL  f.  2. 
Drymaxa  cantanSy  HeugL  Ibis,  1869,  p.  9C. 
Cisticola  cantans,  HeugL  Om.  N.  0.  Afr.  p.  252. 

Iris  orange-brown. 

1  only  met  with  this  species  at  Senafd.  The  tail  is 
imperfect,  and  I  cannot  distinctly  tell  whether  it  has  10 
or  12  rectrices ;  the  latter  is  probably  the  correct  number. 
Wing  2-25,  tail  2*45,  tarsus  0*95,  bill  0*5  in.  I  am 
indebted  to  Dr.  Finsch  for  the  identification  of  this  bird 
with  the  South  African  form. 

This  and  the  two  subsequent  species  belong  to  one 
group,  for  which  Jerdon  (''Birds  of  India," ii.  p.  177)  has 
proposed  the  generic  name  GraminicolcL  It  includes 
several  South  African  species,  and  one  Indian  form, 
G.  hengalensis.  They  are,  I  think,  correctly  distinguished 
from  both  Cisticola  and  Dryiaceca. 

153.  G.  robusta  (lilipp.). 

Drymoica  robusta,  Riipp.  Syst.  Uebers.  p.  35,  t.  xiii. — Lcfebvre, 

p.  89.— Heugl.  Ibis,  1869,  p.  99. 
Cisticola  robusta,  Heugl  Om.  N.  0.  Afr.  p.  254. 

The  only  specimen  obtained  was  killed  at  Adigrat. 
It  has  12  tail-feathers.  Measurement : — AVing  3  in.,  tail 
2-6,  tarsus   I'l,  bill  0*55  in. 

154.  G.  erythrogenys  (Riipp.). 

Cisticola  erythrogenys,  Riipp.  Neu.  Wirb.  p.  111. 

Dryinaxa  erythrogenys,  Riipp.  Syst.  Uebers.  p.  34,  t.  xii. — Ferr.  et 

Gill.  No.  64.— HeugL  Ibis,  1869,  p.  95. 
Cisticola  erythrogenys,  Heugl.  Orn.  N.  0.  Afr.  p.  250. 

Only  a  single  specimen  obtained.  This  also  was  shot  at 
Senafe.  It  measures:  wing  2*5  in.,  tail  2*25,  tjirsus  1,  bill  0*5. 


376  ZOOLOGY. 

Z5S*  Cisticola  habessinica  (Heugl.). 

Hemipterix  hahesiinica,  HeogL  Ibis,  1869,  p,  138 ;  et  Om.  IT.  O. 
Afr.  p.  272. 

Of  this  species  I  obtained  a  single  specimen  only, 
which  was  shot  near  Fokada  in  grass.  Its  habits  and 
flight  resembled  precisely  those  of  the  common  Cisticola 
schcBnicola. 

The  measurements  of  the  specimen  obtained,  a  male, 
are  : — ^Wing  2'11  in.,  tail  1-3,  tarsus  075,  bill  0*4,  whole 
length  about  4*3  in. 

Family  SYLVIADiE, 

156.  Oligocercus  micninis  (Riipp.). 

Troglodytes  mxerurus^  Riipp.  Nen.  Wirb.  p.  109,  t.  xlL  fl  1. 
Oligura  mierura^  Riipp.  Syst  Uebers.  Na  115. 
Sylvietta  braehyurc^  Lefebyre,  p.  89. 
Oligocereus  micruTUg,  Heugl.  Ibis,  1869,  p.  141. 
0.  rufescens,  HeogL  Om.  N.  0.  Afr.  p.  286. 

Iris  orange-brown,  pupil  small ;  legs  deep  flesh-colour ; 
bill  dusky  above,  pale  below. 

A  most  active  and  lively  little  bird,  constantly  himting 
about  bushes  and  trees,  and  often  seen  running  round 
branches,  and  even  hanging  head  downwards.  In  July 
seen  in  pairs.  It  was  only  met  with  in  the  subtropical 
region,  from  about  3,000  to  5,000  feet>  both  in  the  passes 
below  Senafd  and  in  the  Anseba  valley.  It  was  not  rare 
in  those  localities. 

157.  Camaroptera  brevicaudata  (Riipp.). 

Sylvia  hrevicaudatct,  Riipp.  AUas,  t  xxxv.  fig.  b. 
Ficedula  brevicaudata,  Riipp.  Syet  Uebers.  No.  149. 
Camaroptera  brevicattdata,  Brehm,  Habesch,  No.  46. — HengL  Ibis, 

1869,  p.  138  ;  et  Om.  N.  0.  Afr.  p.  281. 
Syncopta  tincta,  Cawin.  Proc.  Ac.  PhiL  1855,  p.  325. 


V  AVES.  377 

^&s  burnt  sienna,  pupil  small,  orbit  pinkish  brown ; 
leg*  flesh  colour ;  bill  black.  In  the  female  the  iris  is 
pei^l  grey,  the  colour  of  the  plumage  duller  and  more 
imiform,  and  the  throat  whiter. 

'I  never  saw  this  bird  in  the  winter,  and  I  am  inclined 
to  suspect  thati  like  GrandcUa  leucogaster,  it  emigrated 
into  the  passes  during  the  spring  montha  In  May,  it 
was  far  from  rare  at  4,000  to  6,000  feet,  and  it  abounded 
on  the  Anseba  in  July  and  August.  The  females  were 
very  rarely  seen,  and  I  only  obtained  one.  The  males 
were  most  conspicuous.  Perched  on  the  topmost  branch 
of  a  tree,  or  some  conspicuous  spray,  they  kept  up  a  re- 
iterated monosyllabic  call  like  "to6k-to6k-to6k,'*  repeated 
at  a  regular  interval.  In  the  stomach  I  found  small 
insects. 

Wing.  TaiL        Tanus. 

in.  in^  in. 

Male 2-3  1-6  1*0 

„ ,     .     .     .     2-26         1-56        0-96 

Female 2*0  1'6  0*8 

From  the  much  shorter  tarsus  in  the  only  female 
obtained,  I  should  have  thought  it  belonged  to  a  distinct 
species,  but  that  it  was  with  a  male  (also  killed).  On 
another  occasion,  when  I  saw  both  together  but  only 
succeeded  in  securing  the  male,  I  noticed  the  paler  tinge 
of  the  female  plumage. 

Ruppell's  figure  and  description  are  probably  taken 
from  a  hen.  C.  tincta  of  Cassen  is  founded  on  a  male 
bird.  Sundevall's  C.  olivacea^  to  judge  from  a  specimen 
in  the  Berlin  Museum,  is  quite  distinct^  being  much 
greener  above. 


378  ZOOLOGY. 

158.  Phylloscopus  abyssinicus,  W.  Blant  (PI.  III.  fig.  2). 

Ph,  Ph.  trochili  similis  sed  supra  magis  viridescens,  tubttu  iiobeUinus 
vix  flavescenSf  cauda  longiorL 

AI.  Caud.         Tars.      Rofitr.  a  fr.    Long,  tota  circa, 

in.  in.  in.  in.  in. 

Mas 2-5        205        077        0*36  45 

Fcemina  ....     2 5        205        079        0*37  4*5 

Common  about  Mayen  in  the  pass  below  Senaf^,  and 
at  Senafd,  Halai,  and  other  places  on  the  highlands.  It 
diflfers  from  P.  Bonelli,  Vieill.  {S.  Nattereri,  Temm.), 
which  it  otherwise  much  resembles,  in  being  pale  buff 
instead  silky  white  below  and  less  brown  above.  From 
P.  trochilus  it  is  distinguished  by  being  more  green  above 
and  less  yellow  below.  It  is  very  close  to  the  Indian 
P.  affinis.  It  may  easily  be  recognised  by  its  isabelline 
coloured  breast. 

The  first  short  primary  is  above  an  inch  (in  one 
specimen  1*1  in.)  shorter  than  the  second,  which  is 
nearly  \  in.  shorter  than  the  third.  The  third  and 
fourth  are  the  same  length,  the  fifth  very  little  shorter. 

I  obtained  several  specimens  of  this  little  Wren- 
warbler  at  Undul  Wells  and  SenafS.  I  believe  it  occurs 
throughout  the  highlands.  I  did  not  meet  with  it  in  the 
Anseba  valley. 

159.  P.  umbrovircns  (Riipp.). 

Ficedula  umbrovirens,  Riipp.  Neu.  Wirb.  p.  112  ;  et  Syst.  Ucbers. 

No.  148.— Ferr.  et  Gal.  No.  71. 
Pkyllopneuste  umbrovirens^  HeugL  Orn.  N.  0.  Afr.  p.  301. 

Only  a  single  specimen  was  obtained ;  this  was  shot  at 
Senafd.  The  wing  measures  2*25  in.,  tail  1*7,  tarsus  0*85. 
The  length  of  the  tail  somewhat  exceeds  that  given  by 
Rlippell. 


AFES.  379 

i6o.  Sylvia  melanocephala,  Gm. 

Syst.  Nat.  i.  pt.  il  p.  970.— HeugL  Om.  N.  0.  Afr.  p.  303. 
Curruca  mdanocephala,  Riipp.  Syst.  Uebers.  No.  131, 

I  shot  a  single  specimen  at  Rairo  in  Habab. 

z6z.  S.  cinerea,  Briss. 

Riipp.  Syst.  Uebers.  No.  133.— Heugl  Om.  N.  0.  Afr.  p.  307. 

Not  rare  in  Samhar  near  Massowa  in  the  middle  of 
August. 

162.  Hypolais  languida,  Hempr.  and  Ehr. 

Curruca  languiday  Hempr.  and  Ehr.,  Symb.  Phys. — HeogL  Om. 
N.  0.  Afr.  p.  21)6. 

From  comparison  with  the  type  specimen  in  the 
Berlin  Museum  I  have  ascertained  that  this  is  identical 
with  Hemprich  and  Ehrenberg's  bird.  1  shot  specimens 
in  the  Lebka  valley  and  Samhar.  It  has  precisely  the 
habits  of  the  Indian  AcrocepJialus  dumetorum  of  Blyth, 
living  amongst  thick  bushes  and  hunting  about  the 
branches  for  insects,  uttering  every  now  and  then  a  sharp 
cry  like  "Ichick-tchick."  It  is  a  restless  active  little 
bird.     The  dimensions  of  a  j)air  are  : — 


Wing. 

TaiL 

Tarsus. 

BiU. 

in. 

in. 

in. 

in. 

Male 

....     30 

2-45 

0-86 

0-53 

Female   .     .     .     . 

....     305 

2-62 

0-85 

0-55 

The  colour  scarcely  differs  perceptibly  from  that  of  the 
next  species.  The  first  primary  extends  about  ^  in.  in 
one  specimen,  ^  in.  in  the  other,  beyond  the  under  wing- 
coverts,  and  is  very  pointed  ;  the  second  is  ^  in.  shorter 
than  the  third,  which  is  the  longest. 


380  ZOOLOGY. 

163.  H.  elaica  (Lindermayer). 

Saliearia  elaica,  Linder.  Ibis,  1843,  pp.  342,  343. 

This  is  a  much  smaller  bird  than  the  last,  easily  dis- 
tinguished by  the  larger  first  primary,  which  is  rounded, 
not  pointed,  at  the  end. 

Measurements  : — Wing  2*6  in.,  tail  2*1,  bill  0*47,  tarsus 
8*5  ;  the  first  primary  is  0*6  in.  long  outside  the  under 
wing-coverts,  and  twice  as  broad  as  in  the  last  species ; 
second  primary  0*2  in.  shorter  than  the  third,  which  is 
the  longest. 

r64.  Aedon  galactodes,  Temm. 

Sylvia  galactodes,  Temm.  PI.  CoL  251,  £  1. 

A,  minor.  Cab.  Mus.  Hein.  L  p.  39. — Brehm,  Habesch,  No.  48. 

Aedcm  galactodes,  HeugL  Ibis,  1869,  p,  84 ;  et  Oip.  N,  0.  Jifr^  p.  276. 

Common  fi-oni  the  base  of  the  hills  to  about  3,500  or 
4,000  feet,  but  not  seen  above.  In  the  winter  (January 
and  February)  it  was  only  seen  at  the  base  of  the  ranges  ; 
in  May  it  was  comjnon  as  high  as  Mayen,  about  3,500 
feet.     It  appears  to  be  confined  to  the  tropical  region. 

Family  ALAUDIDiE. 
165.  Motacilla  alba,  L. 

Riipp.  Syst.  Uebers.  No.  173.— HeugL  Cm.  N.  0.  Afr.  p.  316. 
?  M.  dukhunmsis,  Sykes,  P.  Z.  S.  1832,  p.  91. 

Common  both  on  the  highlands  and  near  the  coast.  I 
cannot  see  that  the  specimens  collected  by  me  in  Abys- 
sinia show  the  slightest  difference  from  others  shot  in 
Western  India,  and  belonging  to  the  supposed  species 
M.  dukkunensis,  Sykes. 

On  the  1st  of  May  there  were  still  specimens  on  the 


AVES.  381 

highlands  around  Lake  Ashangi,  but  only  very  few  re- 
mained.    A  month  earlier  they  had  been  numerous. 

z66.  M.  (Calobates)  sulphurea  (Bechsi). 

MotaciUa  sulphurea^  HengL  Om.  N.  0.  Afr.  p.  318. 

Not   common.     I  only  killed  one  specimen,  which  I 
\  obtained  in  the  Lebka  valley  in  August. 

167.  Budytes  melanocephala   (Licht.),  or  B.  flava^  L.  yar. 
mdanocepluila. 

MotaciUa  flava  mdanocephdla,  HeagL  Om.  N.  0.  Afr.  p.  321. 
Mota4:iUa  mdanocepKalay  licht.  Veiz.  der  DoubL  p.  36. — Riipp. 
Atlas,  t.  xxxiil  b. ;  et  Syst  Uebers.  No.  178. 

Common  everywhere  during  the  winter,  and  I  suspect 
many  remain  and  breed  on  the  highlands  of  Abyssinia, 
for  birds  of  this  species  were  still  abundant  around  Lake 
Ashangi  at  the  beginning  of  May,  although  they  had 
then  assumed  the  nuptial  plumage  more  than  a  month. 

The  Yellow  Wagtails  pass  gradually  into  each  other, 
and  are  considered  by  many  of  the  best  omithologiste,  as 
Tristram  and  Finsch,  as  merely  varieties  of  one  species, 
B.Jlava,  L. 

Specimens  of  B.  melanocephala  killed  in  December 
and  January  are  perfectly  indistinguishable  from  the 
Indian  specimens  generally  assigned  by  Jerdon  and 
others  to  B.  viridis,  Gm. 

z68.  B.  cinereocapilla,  Savi 

?  B.  flava^  L.  Riipp.  Syst.  Uebers.  No.  176. 
MotacUlaflawi  cinereoscapilla,  HeugL  N.  0.  Afr.  p.  321. 

A  single  specimen  in  breeding  plumage  was  shot  at 
Lake  Ashangi  at  the  beginning  of  April     It  is  interme- 


382  ZOOLOQT. 

diate  between  the  var.  cinereoscapilla  and  true^iw,  as 
it  has  only  a  trace  of  a  white  superciliary  stripe. 

169.  B.  campestris,  Pallas,  ot  B,Jlava^  L.  yar.  campestrisl 

A  single  specimen  obtained  on  the  highlands  in  spring 
has  the  forehead  and  greater  portion  of  the  crown  of  the 
head  yellow.  Mr.  Tristram  informs  me  that  he  has  seen 
but  one  other  specimen  with  an  equally  yellow  head. 
That  was  shot  in  England.  I  am  informed  by  Dr.  Finsch 
that  this  race  is  the  true  Motacilla  campestris  of  Pallas, 
and  that  M.  Rayi  of  Linnaeus  diflfers  by  having  the 
crown  greenish  and  well-marked  yellow  supercilia.  Mr. 
Tristram,  however,  does  not  consider  that  this  is  Pallas's 
bird.  I  confess  myself  unable  to  decide  anything  with 
respect  to  these  races. 

170.  Anthus  cecilii,  Savigny. 

Say.  Descr,  de  IHSgypte,  Explic  somin.  des  Planches,  p.  281. — Ois. 

pL  5, 1  6, 1826. 
Motacilla  cennna,  Pallas,  Zoogr.  Rob.  As.  L  p.  511, 1831. 
Anthus  eervinuSf  Heugl.  Orn.  N.  0.  Afr.  p.  323. 

Common  around  Senaf^  on  ploughed  fields. 

171.  Anthus  {Agrodroma)  sordidus,  Eiipp.  or  A.  campestris, 
yar.  sordtdits. 

Anthus  sordidus,  Riipp.  Neu.  Wirb.  p.  103,  t.  xxxix.  f.  1. — Heugl. 
Om.  N.  0.  Afr.  p.  326. 

This  bird  varies  much,  apparently  with  age  and  the 
state  of  the  plumage.  Some  specimens  are  almost  uni- 
formly dusky  above,  with  part  of  the  outer  tail-feathers 
alone  pale  buff,  and  only  a  trace  of  pale  edgings  to  the 
quills  and  wing-coverts.  This  is  the  plumage  of  two 
specimens  shot  in  the  Anseba  valley  in  July.     All  the 


AFES. 


383 


specimens  shot  on  the  highlands  in  the  winter  and  spring 
are  more  rufous,  and  have  more  or  less  broad  rufous 
margins  to  the  taU-feathers,  quills,  and  wing-coverts. 
The  greater  portion  of  the  outer  tail-feathers  and  a  por- 
tion of  the  next  pair  are  whitish  in  colour,  the  centres  of 
the  latter  being  usually  dusky.  The  coloration  of  the 
underparts  also  varies  slightly.  In  sonie  birds  the  breast 
is  distinctly  spotted ;  in  others  there  is  not  a  trace  of 
spots.  Mr.  Tristram,  who  has  kindly  examined  my 
specimens,  considers  that  all  the  varieties  belong  to  one 
species,  and  I  am  quite  disposed  to  agree.  There  is  great 
variation  in  dimensions.  I  give  those  of  all  the  specimens 
collected,  five  in  number : — 


Wing. 

Tail. 

Tarsus. 

(about) 

BilL 

in. 

in. 

in. 

in. 

in. 

Anseba    . 

.     3-8 

3-2 

1*05 

6-76 

0-65 

Male,  dnsky. 

>> 

3-5 

3-0 

1-0 

6-5 

0-6 

Female,   „ 

Senaf^ 

.     4-0 

3-4 

1-0 

7-0 

0-56 

Male,       „ 

» 

.     3-9 

3-2 

1-05 

7-0 

0-05 

Female,more  rufous. 

» 

.     3-55 

2-5 

105 

575 

0-56 

Young,  very  pale. 

It  is  just  possible  that  the  last  specimen,  which  is  entirely 
without  spots  on  the  breast,  should  be  referred  to  A. 
camjyestris,  L.,  but  I  am  disposed  to  consider  it  a  young 
specimen  of  A,  sordidus. 

Anthiis  sordidus  keeps  much  to  rocky  hill-sides,  where 
there  is  but  little  bush.  It  was  common  near  Seiiafd, 
and  equally  so  on  the  more  barren  hills  bordering  the 
Anseba  valley.     It  sometimes  perches  on  trees. 

172.  A.  {Agrodroma)  campestris  1 

Antlius  campestris,  Heugl.  Orn.  N.  0.  Afr.  p.  325. 

Only  seen  on  the  highlands.    There  it  was  everywhere 


384  ZOOLOGY. 

abundant  on  grassy  meadows.    A.  cervinus  appeared  to 
replace  it  on  cultivated  land. 

It  is  quite  impossible  in  the  series  preserved  by  me  to 
distinguisli  this  species  from  the  last.  As  a  rule  A.  cam- 
pestris  is  a  smaller  bird,  brighter  and  more  rufous  in 
colour  above,  and  paler  beneath,  with  a  distinct  spotted 
breast.  The  bill  is  conspicuously  smaller.  But  not  one 
of  these  characters  can  be  relied  upon.  I  append 
measurements  of  specimens,  by  comparing  which  with 
those  of  the  last  species  it  will  be  seen  that  there  is 
complete  gradation,  and  the  colour  also  varies : — 

Wing.         TaO.         Taraiu.  BIIL 

in.  in.  tn.  in. 

Female 3*75  27  I'l  0*53  Seiiaf(6 

Male 3-65  2*65  11  0*57  Adigrat 

Female 3*5  2*65  1*07  0*6  Ashangi 

„        3-4  2-6  1-05  0-64  „ 

The  first  two  should  perhaps  rather  be  classed  in 
A.  sordidtis,  but  it  is  simply  impossible  to  draw  a 
distinct  line. 

173.  Macronyx  flavicoUis,  Kiipp. 

Riipp.  Neu.  Wirb.  t.  xxxviiL  fig.  2.— HeugL  Cm.  N.  0.  Afr.  p.  330. 

Iris  brown ;  bill  black  above,  bluish  grey  below ;  legs 
brown. 

In  habits  and  flight  this  bird  is  rather  a  Pipit  than  a 
Lark,  and  it  wante  the  rudimentary  first  primary  quill, 
though  its  biU  and  feet  are  Lark-like.  It  was  only  seen 
on  the  highest  part  of  the  Dalanta  plateau,  and  near  the 
crest  of  the  Wandaj  pass,  and  was  never  observed  below 
10,000  feet.  It  was  not  common,  but  occurred  generally 
in  pairs  ;  sometimes  three  were  seen  together. 


AFES.  385 

•  Measurements  taken  in  the  flesh :  whole  length  6 '75 
in.,  closed  wing  3'85,  tail  2'12,  tarsus  1*12,  hind  toe 
0*5,  claw  0'3,  bill  in  front  05  in. 

174.  Certhilauda  desertorum  (Stanley). 

Alauda  deseriorwm,  Btanley,  Salt's  Joamey,  App.  p,  Ix. 
CerihiloAJLda  districrumj  Riipp.  Syst  Uebera.  No.  304. 
Aktman  detefiorumy  HeugL  Joxan.  L  Om.  1868,  p.  S30. 

Rather  common  about  Zulla,  but  comparatively  rare 
near  Massowa,  The  flight  of  this  bird  did  not  appear  to 
me  at  all  Plover-like;  it  was  certainly  not  the  least 
similar  to  that  of  other  Larks ;  but  although  more  rapid, 
there  was  something  in  the  mode  of  flight  which  always 
reminded  me  of  a  Hoopoe,  and  the  two  conspicuous 
white  wing  bands  greatly  increased  the  resemblance.  It 
is  not  a  wary  bird,  but  when  running  it  is  very  swift.  It 
keeps  to  comparatively  open  spots,  and  is  usually  solitary 
or  in  pairs.  I  only  once  saw  three  together,  and  then 
two  appeared  to  be  young  birda 

The  bill  varies  much  in  length,  and  appears  to  be 
longest  in  the  male.  I  append  measurements  of  four 
specimens : — 

Wing.         TaU.         BUI.    .    Tamu. 
in.  in.  in.  in. 

1.  Male 6-0  3*8        I'l  1*3 

2.  „ 4-8  3-6         1*0  122 

3.  Female 4*3  3'2        0*85        1*22 

4.  „ 4'26        3-0        0-86         13 

Much  doubt  appears  to  overshroud  the  synonomy  of 
this  bird.  It  was  originally  described  by  Lord  Stanley, 
from  a  specimen  brought  by  Mr.  Salt  from  Amphila,  on 
the  Abyssinian  coast,  and  was  characterised  as  greyish 
sandy  brown,   A  more  rufous  race  inhabiting  Nubia,  which 

CO 


386  ZOOLOGY, 

passes  into  the  grey  one  by  imperceptible  degrees,  is  the 
type  of  A.  hifasciata,  Licht.  I  do  not  intend  to  enter 
at  length  into  the  question  of  the  identity  or  distinctness 
of  these  two  races,  but  merely  to  show  that  if  they  are 
to  be  distinguished,  the  name  desertorum  must  be  re- 
tained for  the  greyer  form,  found  on  the  shores  of  the 
Red  Sea  and  in  Arabia,  and  hifasciata  for  the  more 
rufous  race  which  is  found  in  Nubia.  Latham's  Desert 
Lark  ("  Gen.  Hist."  vol.  vL  p.  301),  as  was  pointed  out  to 
me  by  Dr.  Finsch,  seems  to  apply  to  the  hifasciata  type, 
as  the  head,  neck,  and  back  are  said  to  be  dun  or  brownish 
buflF  colour,  and  the  specimen,  although  stated  to  be  in 
the  collection  of  Mr.  Salt,  can  scarcely  have  been  the  same 
as  the  type  described  by  Lord  Stanley.  The  synonymy 
will  be : — 

Certhilauda  (Alauda)  desertorum,  Stanley,  loc  cit. — HeugL 
No.  183.— Faun,  d  Bothen  Meer.— Petennan's  Mitth.  1861. 
—Ibis,  1859,  p.  343.— Jo\im.  t  Cm.  1865,  p.  230. 

0.  bifcudata,  Gk>uld,  B.  Europe,  pL  168. 

(This  is  a  somewhat  intermediate  variety.) 

C  Salmniy  Tristram,  Ibis,  1859,  p.  57. 

Certhilauda  (Alauda)  bifasciata,  Lioht.  Yerz.  d.  DoubL  p.  27. 

— Riipp.  Atlas,  t.  v.— Temm.  PL  CoL  393. 
C,  duertorum,  Stanley,  apnd  Latham,  Qen.  Hist.  tL  p.  301. — Riipp. 

Syst.  Uebers.  p.  78,  No.  304.— HeugL  Syst.  Uebers.  No.  438. — 

?  Jerdon,  Birds  of  India,  iL  p.  438. 
?  Saxieola  IpaUidOy  Blyth,  Jouiil  As.  Soc.  Bengal,  xvi  p.  130. 

(This  also  appears  somewhat  intermediate,  and 
Jerdon  notices  that  an  Indian  specimen  is 
darker  than  that  figured  by  RiippelL) 

My  specimens,  all  from  the  same  locality,  show  great 
Variation  in  the  amount  of  spotting  on  the  breast,  though 


AFES.  387 

all  have  far  more  than  Nubian  specimens.  The  relative 
extent  of  the  two  white  bands,  and  the  intervening 
dusky  space  on  the  secondary  quills,  are  also  marvel- 
lously inconstant,  and  even  the  colour  of  the  back  varies 
somewhat  Some  specimens  from  Arabia  in  the  Berlin 
Museum  have  the  breast  even  darker  and  more  thickly 
spotted  than  in  the  birds  from  Annesley  Bay.  There  are 
1  amongst  the  ve^  fine  «riea  in  t  Jm  Jlun.  dedded 
traces  of  a  passage  from  the  grey  C  desertorum  to  the 
rufous  (7.  hifdsciatay  and  as  none  of  the  characters  which 
at  first  sight  distinguished  the  two  appear  to  be  at  all 
constant,  even  in  birds  from  the  same  locality,  I  should 
be  disposed  to  unite  the  two  races. 

175.  Alauda  {GaUrita)  cristata,  L. 

O,  lutea,  Brehm,  Habesch,  No.  97. — ^Riipp.  Syst.  Uebers.  No.  309. 

— HeugL  Journ.  f.  Orn.  1868,  p.  233. 
?  Gdlerida  isahellina,  Bon.  Consp.  p.  242. 

Common  near  the  coast.  I  can  see  no  difference 
between  specimens  from  the  Red  Sea  and  others  from 
India,  and  do  not  understand  why  Bonaparte  proposed 
to  separate  them. 

I  obtained  at  Zulla  an  albino  specimen  of  this  bird  of 
pale  isabelline  colour,  without  any  dark  marks  either  on 
the  back  or  breast 

176.  A.  {G,)  arenicolal  Tristram,  var. /i^co. 

Ibis,  1869,  p.  58. 

I  obtained  a  single  specimen  of  a  very  dark-coloured 
Lark  on  the  Abyssinian  highlands  at  Ashangi,  which  I 

cc  2 


388  ZOOLOGY. 

was  inclined  to  refer  to  G.  ahyssinica,  Bon.,  althougli  its 
dimensions  are  much  smaller. 


Wing 

TaU. 

BilL 

Tannu. 

in. 

in. 

in. 

in. 

4 

2-3 

0-65 

105 

Mr.  Tristram,  however,  informs  me  that  it  agrees  in 
everything  except  colour  with  his  G.  arenicola. 

177.  Alauda  prsetermissa,  W.  B]an£  (Plate  YL). 

Ann.  and  Mag.  Nat.  Hist.  Not.  1869. 

A,  iupra  fuico-umbrinaf  capitis,  coUi  postid,  irUeneapuliique 
plumis  late  et  paUide  rufeseenti  mar^natis^  dorso  pasteriori 
magis  cirurascenH,  vix  striolaJto ;  iupereUiis  et  gadreo  UAo 
itaheUinis;  geni$f  colli  lateribttSf  pecioreque  iaJturaJt%orib%u  et 
fusco-guttatis,  regione  pwroticd  fusceecenti;  remigihus  et  teetri- 
cihuu  alarum  umbriniey  primariie  extue  ieabellino  tecundarite 
teetricibusque  rufo-marginaii$y  remigihiLS  omnibus  intus  vemu 
basin  rufo-fulvis ;  uropygio  et  reetricibus  m^ianis  nrfescenii- 
umJbriniSy  reetricibus  exteris  pogoniointemo  basin  versus  exceptis 
et  pagoniis  extemis  secundarum  rufo-isabellinisj  eceteris  cum 
partibus  rdiquis  extemarum  fumoso-nigricantibus.  Caput  suh- 
crisUUumf  rostrum  supra  fuscum,  subtus  paUidum,  pedes  eamei 
irides  brunnete. 

Long.  alB.    Gaud.       Tart.  Ung.  post  Rost  a  fr.  Tota  circa, 
in.  in.  in.  in.  in.  in. 

Dimensiones  avis  maris  .    3'9        2*16       10      0*45         0*5  6*25 

„  „  fceminfiB    3*7        2*0         1*0       0*4  0*5  6*0 

The  first  primary  is  J  to  tV  inch  in  length,  the  second 
very  little  shorter  than  the  third,  which  is  nearly  equal 
to  the  fourth. 

I  found  this  to  be  the  common  Skylark  around  Senaf^ 
and  Adigrat,  with  precisely  the  habits  of  the  European 
bird.  It  is  distinguished  by  the  bill  being  stronger  and 
the  wing  decidedly  shorter.  I  cannot  find  any  descrip- 
tion agreeing  with  it  amongst  the  birds  included  in 
Von  Heuglin's  lists ;  and  as  Mr.  Gray,  Dr.  Finsch,  and 


I 

I 


JVE8.  389 

Mons.  Jules  Verreaux  aU  agreed  with  me  in  considering  it 
new,  I  described  it  as  such.  Mr.  Tristram  considers  it  a 
fonn  of  Galerita  cristata,  in  which  I  cannot  coincide ;  the 
nearest  described  species  is  probably  Alanda  erythropygia 
of  Strickland  (P.  Z.  S.  1850,  Aves,  pi.  xxiv.),  but  the 
present  form  is  smaller,  and  the  rufous  colouring  on  the 
tail-feathers  is  differently  distributed. 

Z78.  Calandrella  brachydactyla  (Temm.). 

Riipp.  Syst  Uebers.  No.  306. — ^HeugL  JoarxL  £  Orn.  p.  232. 

Not  rare  on  the  shores  of  Annesley  Bay.  I  did  not 
meet  with  it  in  the  highlands. 

Z79.  C.  Anderssoni  (Tristram).  KS.  Ibis,  1869,  p.  434. 

Closely  allied  to  A.  cinerea^  Latham  {A.  rujiceps^ 
Riipp.),  but  smaller,  and  without  the  black  spot  at  the 
side  of  the  neck.  Mr.  Tristram  informs  me  that  he 
considers  it  distinct,  and  has  named  it  after  the 
South  African  Naturalist,  who  discovered  it  also  in 
Damara  Land. 

This  bird  is  unquestionably  a  Calandrella,  and  per- 
fectly congeneric  with  the  last  species,  which  it  probably 
replaces  on  the  Abyssinian  highlands,  both  in  form  and 
habits. 

I  found  this  small  Lark  abundant  on  stony  ground 
near  Senaf^,  keeping  much  in  large  communities,  and 
highly  social,  precisely  like  (7.  brachydactyla. 

A  male  specimen  measures:  wing  3*25  in.,  tail  2*1, 
tarsus  0*8,  hind  claw  0*35,  bill  from  front  0*4  in. 


390  ZOOLOGY. 

i8o.  Ammomanes  deserti  (licht.). 

Alauda  deserti,  lichtenstein,  Yerz.  d.  DoabL  p.  28. 
A.  isabeUinOy  Temm.  PL  CoL  144,  t  2. 
Melanocorypha  ieabeUiruij  Riipp.  Syst.  Uebeis.  No.  306. 
Ammomanei  deserti,  Hengl.  Jouro.  f.  Orn.  1868,  p.  226. 

Not  rare  to  the  west  and  north-west  of  Massowa, 
in  rocky  and  stony  ground,  amongst  thin  bushes. 

A  specimen  measures:  wing  3*7  in.,  tail  2*5,  tarsus 
8'7,  biQ  fix)m  nostril  to  point  0*42  in. 


i8z.  Pyrrhulauda  melanauchen,  Cab. 

CoraphHes  melanauchen,  Cabanis,  Mas.  Hein.  L  p.  124. — HetigL 

Journ.  f.  Om.  1868,  p.  219. 
Pyrrhulauda  crucigera,  Riipp.  Syst.  Uebera.  No.  313. — Brebm, 

Habesch,  No.  101. 

Extremely  abundant  in  all  the  sandy  country  near  the 
coast.  These  birds,  like  other  species  of  the  genus,  are 
true  Larks  and  not  Finches  in  all  their  habits:  they  always 
live  on  the  ground,  very  rarely  indeed,  if  ever,  perching 
on  bushes;  they  are  social,  several  keeping  about  one  spot, 
but  they  do  not  exactly  associate  in  flocks,  and  when 
disturbed  rise  one  after  the  other,  not  simultaneously, 
and  fly  ofi*  without  keeping  close  to  each  other.  In  the 
middle  of  the  day  several  may  be  seen  together  in  the 
shade  of  a  tree  or  a  bush.  They  run,  too,  and  do  not 
hop,  and  their  flight  is  lark-like.  Altogether,  con- 
sidering their  flight,  habits,  and  plumage,  there  can  be 
no  question  that  they  are  true  Larks  and  not  Finches, 
and  the  presence  of  a  small  first  primary  confirms  this 
position. 

The  following  are  the  measurements  of  a  pair : — 


AFES.  391 


Wing. 

TkiL 

^^t."   T«.UB. 

in. 

in. 

in.           in. 

Male,  Zulla,  Annesley  Bay  .    . 

3-05 

1-96 

0-4         07 

Female,  Massowa 

31 

1-95 

0-41       075 

The  outer  rectrices  in  both  sexes  are  isabelline  nearly 
throughout ;  the  margins  of  the  second  pair  and  the  tips 
of  the  remainder  are  the  same,  the  colour  decreasing 
towards  the  middle. 

i82.  P.  albifrons  (Sund.). 

CoraphUa  albifrons^  Sund.  Offers,  k.  Vet.  Akad.  Forb.  1850, 

p.  127. 
"  P.  frontalis^  licht,**  Bon.  Consp.  Gen.  Av.  L  512. 
C,  nigriceps,  Heugl  Joum.  1  Om.  1868,  p.  218. 

Some  of  the  specimens  obtained  by  me  appear  to 
belong  to  this'  species,  from  which  the  last  is  but  doubt- 
fully distinct.  I  cannot  find  the  original  description  by 
Lichtenstein,  but  the  species  is  by  Cabanis  ("Joum.  f. 
Om."  1868,  p.  219,  note)  said  to  be  distinct  from 
C.  nigricepSy  Gould,  This  is  probably  the  case,  judging 
from  Gould's  description  (*'  Voy.  Beagle,"  Birds,  p.  87),  in 
which  the  C.  nigriceps  from  the  Cape  Verde  Islands  is 
said  to  have  the  centre  of  the  back-feathers  dark,  and 
a  white  line  on  each  side  of  the  breast,  neither  of 
which  characters  are  found  in  the  East  African  form: 
the  dimensions  of  the  latter  are  also  slightly  larger, 
Sundevall's  name,  dated  May  1850,  has  priority  over 
Lichtcnstein's  if  the  latter  was  not  published  before 
given  to  the  world  in  Bonaparte's  "  Conspectus,''  as  the 
portion  containing  it  bears  date,  August  1850.  The 
name  does  not  occur  in  the  "  Verz.  d.  Doubletten." 


392  ZOOLOGY. 

As  I  did  not  discriminate  between  the  two  races  amongst 
the  numerous  specimens  of  this  bird  which  I  brought  from 
Abyssinia,  I  am  inclined  to  suspect  that  they  pass  into 
each  other.  In  the  only  adult  male  specimen  now  before 
me,  which  I  refer  to  P.  frontalisy  the  ears  are  grey  and 
the  cheek-patch  not  separated  by  a  black  line  from  the 
grey  of  the  back.  In  a  younger  specimen  the  ears  are 
white ;  the  white  of  the  forehead  is  much  less  extended. 
In  both  birds  the  rectrices  have  no  pale  margins.  In  a 
female,  which  I  refer  also  to  this  species,  the  outer  rectrices 
have  the  greater  portion  of  inner  web  blackish  brown  ; 
the  outer  web,  with  part  of  the  inner,  and  the  outer 
edges  of  the  second  pair,  are  isabelline.  The  following 
are  the  measurements  : — 

Wing.  TftiL 

Male,  Zulla 3*6  1*95 

Da  young,  Amba,  Samhar       3*0         1*9 
Female,  Zolla 2*95        1*8 


BiU. 
in. 

0*37 

Tamuk 

in. 
066 

0*4 

0*iB2 

0*36 

0*67 

183.  p.  leucotis  (Stanley). 

Zroxia  UucoUsy  Stanley,  Salt's  Journey,  App.  p.  Ix. 
Pyrrhulavda,  UvMtis,  Riipp.  Syst.  Uebers.  No.  314. 
CoraphiteB  leucoti$y  HeugL  Joum.  t  Cm.  1868,  p.  217. 

Iris  raw  sienna ;  beak  white ;  legs  flesh  colour. 

This  Lark  inhabits  the  same  localities  as  the  last,  and 
associates  with  it,  so  that  I  have  killed  specimens  of  both 
with  one  shot  It  is,  however,  far  less  abundant.   Dimen- 


sions : — 

Male 

Wing, 
in. 
.     .      .     .        3*0 

TaH. 
in. 
1*9 

20 

Bill. 

in. 

0-35 

0-35 

Tarsus, 
in. 
0*6 

Female     .... 

....       31 

0-66 

AFEs.  39a 

Family  COKVIDiE. 

184.  Corvus  afiinis,  Biipp. 

Neu.  Wirb.  p.  20,  t  x.  1  2 ;  Syst  Ueben.  No.  239.— Tristram, 

Ibis,  1866,  p.  72.— Heugl.  Joum.  1  Om.  1868,  p.  316. 
Corax  affinis,  Brehm,  Habesch,  No.  79. 

Iris  dark  brown ;  bill  and  legs  black. 

This  Crow  has  been  well  described  by  Tristram  (loc.  cit.). 
In  the  air  it  may  immediately  be  recognised  by  its  short  tail 
and  long  secondary  quills,  and  its  croak  is  much  deeper 
than  that  of  most  Crows.  It  abounds  everywhere  on  the 
highlands  and  in  the  subtropical  zone,  and  it  descends 
almost  to  the  sea-level  at  times.  When  I  first  visited 
Komayli,  at  the  base  of  the  hills,  in  January,  the  only 
Crow  to  be  seen  was  C.  scapulatu^,  but  in  February,  after 
some  rain  had  fallen,  C.  affinis  abounded.  In  the  former 
instance  I  first  met  with  this  species  at  about  3,000  feet, 
near  Mayen.  In  May  it  had  retreated  once  more  to  its 
former  range,  and  the  white-breasted  Crow  alone  was  to 
be  seen  in  the  tropical  region.  In  ascending  to  the 
Bogos  country  it  first  appeared  on  the  Lebka  at  about 
12,000  feet  above  the  sea,  at  Ain, 

185.  C.  scapulatus,  Daud. 

Daudin,  Traits  complet  d*Om.  u,  p.  232. — Lefebvie,  p.   105. — 
Biipp.  Syst.  Uebers.  No.  238.— HeugL  Joum.  f.  Orn.  p.  313. 

The  distribution  of  this  species  is  most  remarkable. 
It  was  common  near  the  coast,  and  was  always  found  in 
smaller  or  larger  numbers  about  Zulla,  Komayli,  &c.,  but 
not  further  up  the  passes.  In  ascending  the  Lebka 
valley,  N.W.  of  Massowa,  it  only  appeared  as  far  as 
about  3,000  feet  above  the  sea,  near  Kelamet.    Through- 


394  ZOOLOGY. 

out  Tigr6  and  in  the  Anseba  country  I  never  once  saw  it 
When,  therefore,  I  came  upon  a  Crow  with  a  white  collar 
on  the  top  of  the  Wadela  and  Dalanta  plateaux,  close  to 
Magdala,  at  an  elevation  exceeding  9,000  feet,  I  confi- 
dently expected  that  it  must  prove  a  distinct  species^ 
and  it  was  only  after  most  careful  comparison  that  I 
became  satisfied  of  its  perfect  identity  with  the  form  so 
common  on  the  coast. 

This  Crow  has  a  harsh  deep  croak,  but  less  deep  than 
that  of  C.  affinis. 

are    measurements  taken  from   four 

BiUto 

gap6. 

in. 

2-3 
2-4 
2-3 
2-15 

i86.  Corvultur  crassirostris  (Eiipp.). 

Baveu,  Brace,  vol.  v.  p.  153. 
Corvus  crasdrostris,  Riipp.  Neu.  Wirb.  p.  18,  t.  viiL 
C,  dlbicollis,  Lefebyre,  p.  104. 

Corvultur  crasdrostrisy  Eiipp.  Syst.  Uebers.   No.   142.  —  Heugl 
Joum.  1  Om.  1868,  p.  317. 

Iris  very  dark  brown  ;  bill  and  legs  black. 

Common  in  places  on  the  higher  portions  of  the  high- 
lands, but  locally  distributed.  I  first  saw  it  at  Fokada, 
two  marches  south  of  Senafd,  again  about  Dolo  and  other 
camps  a  little  north  of  Antalo,  and  again  abundantly  in 
the  Wadela  plateau.  At  Yasendye,  three  marches  from 
Magdala,  all  three  species  of  Crows  occurred. 

Corvultur  crassirost7%s  has  a  deep  guttural  croak,  and 
appears  to  be  more  exclusively  carrion-feeding  in   its 


The    following 

are    me 

specimens : — 

Length 

Male,  Habat    . 

In. 
.     .       21 

„     Zulla .    . 

.     .       21 

„     Wadela  . 

.     .       20 

Female     ,, 

.     .       18 

Wing. 

Tail. 

Tarsus. 

in. 

in. 

in. 

14-5 

8-0 

2-6 

14-26 

8-0 

2-6 

16-0 

9-4 

2-6 

14-26 

776 

2-6 

AFES.  395 

habits  than  other  species.  I  always  saw  it  aflsociating 
with  Vultures  near  carcases.  It  flies  heavily,  expanding 
its  tail  much.  Other  Crows  {C.  affinis)  pursue  and  tease 
it,  just  as  they  do  Lammergeyers,  Eagles,  and  other  birds 
of  prey. 

Fregilits  graculus  I  met  with  once  on  the  Wadela 
plateau,  near  Gaso,  at  an  elevation  of  10,500  feet,  but  I 
did  not  secure  a  specimen.  There  was  a  small  flock,  not 
very  wary;  indeed  I  approached  so  near  that  I  could 
only  account  for  my  not  killing  a  bird  by  my  cartridges 
being  damp.  They  appeared  to  be  searching  for  insects 
on  the  ground. 

Family  STUENIDiE. 
187.  Lamprocolius  chalybseus  (Hemp,  and  Ehr.). 

Lamprotomis  chcUybasus,  Hemp,  and  Ehr.,  Sym.  Phys.  t  x. — Riipp. 

Syst.  Uebers.  No.  242.— HeugL  Joum.  1  Om.  1869,  p.  5. 
X.  cyaniventris,  Blyth,  Journ.  As,  Soc.  Bengal,  1855,  p.  255. 
L.  chalyhcetu  and  L,  ahyssinicuSy  Hartlaub,  Joum.  £  Om.  1859, 
p.  21. 

Iris  golden  yellow,  sometimes  orange  or  scarlet  near 
the  pupil. 

This  bird  varies  greatly  in  size,  so  much  so  that  it 
is  not  at  all  surprising  that  two  races  should  have 
been  distinguished.  But  despite  the  great  differences 
in  the  specimens  obtained,  amounting  to  1^  inches  in  the 
length  of  the  closed  wing,  I  am  unable  to  see  any 
constant  distinction,  there  being  numerous  specimens  of 
intermediate  size.  The  following  series  of  measurements 
will  illustrate  this  : — 


398  zooLOGr. 

Wing.  Tail.  Billatfhmt  Tarsns. 

in.  in.  in.  in. 

1.  Haik-hallat,  near  Antalo     .       5*9  4*2        075         1'25 

2.  Eokai,  Lebka  vaUey,  Hahab      5*8  4*0        0*7  1*3 
a  Musga,  near  Antalo   ...      5*3          3*5        0*7  1*2 

4.  Lake  Ashangi 5*15        3*5        0*6  1*3 

5.  Anaeba  valley 475        3*0        0*6  1*3 

6.  Adat,  Samhar 4*6  3*2        0*6  115 

7.  „      (a  young  bird)      .    .      4*2  1*6        0*65        1*1 

It  is  very  difficult,  if  not  impossible,  to  draw  a  line 
here.  The  type  specimens  of  L.  cyaniventris}  Blyth,  in 
the  Indian  Museum  at  Calcutta,  which  I  have  examined, 
belong  to  the  larger  race,  to  which  indeed  all,  or  almost  all, 
of  the  numerous  specimens  obtaiiied  on  the  table-land 
may  be  referred,  the  skin  from  Lake  Ashangi  (No.  4  of 
the  above  table)  beiijg  the  smallest  preserved;  whilst  the 
few  specimens  from  the  low  country  are  much  smaller. 
Both  varieties  were  found  on  the  Anseba.  The  iris  in 
the  large  variety  appeared  to  be  more  orange,  but  this  is 
very  possibly  a  difference  depending  upon  age. 

Lamprocolius  chalybcms  was  not  seen  in  the  passes  or 
at  Senaf6  in  January,  February,  or  March.  It  was  occa- 
sionally met  with,  as  at  Fokada,  Dongolo,  and  Lake 
Ashangi,  on  the  march  to  Magdala,  but  usually  at  this 
season  appeared  confined  to  lower  levels.  In  May,  how- 
ever, it  was  abundant  about  Senaf^  and  throughout  the 
upper  part  of  the  passes,  and  it  was  also  met  with 
commonly  in  July  in  the  Anseba  valley,  and  even  at  the 
base  of  the  hiUs  near  Ailat,  but  it  was  not  observed  in  the 

*  This  name  is  prior  by  four  years  to  that  of  L.  ahymnicm ;  and  if  the 
two  races  be  kept  distinct,  as  is  done  by  Von  Ileuglin,  must  be  adopted 
for  the  larger  one. 


AVES.  397 

coast  region.     I  obtained  birds  in  the  young  plumage 
figured  by  Rtippell  in  July. 

This  bird  has  very  much  the  flight  and  habits  of  the 
common  Starling. 

i88.  L.  chrysogaster  (Gm.). 

Turdu9  chrysogaster,  Qm.  Svst  Nat.  i.  pt.  il  p.  835. 
Lamprotomis  rufiverUris,    Riipp.   Neu.  Wirb.  p.  27,  t.  xL  £  1 ; 

•   Syst  Uebeis.  No.  247. 
Nota/uges   ehrysogasUr,  Cab.  Mob.   Hein.   L  p.   198.  —  HarUaab, 
Joam.  I  Ora.  1869,  p.  26.  —  Heugl  Jouzn.  t  Om.  1869,  p.   9. 

Iris  yellowish  white  ;  legs  and  bill  black. 

This  species  was  never  observed  on  the  highlands  or  to 
the  south  of  Massowa,  but  it  was  common  about  Ailat, 
Ain,  &c.  at  the  base  of  the  hills,  and  in  the  Lebka 
and  Anseba  valleys,  usually  occurring  in  small  companies 
or  flocks,  and  feeding  on  the  ground  on  insects,  not 
unfrequently  in  company  with  L.  chalybcBus,  which  it 
greatly  resembles  both  in  flight  and  habits.  Measure- 
ments : — 

Wing.         TaiL  BilL         Tareoji. 

in.  in.  in.  in. 

Male  from  Ailat 4*5        2-86        075        1*25 

Female,  Anseba  valley  .     .    .      4*4        2*9  0*65        1*25 

189.  Lamprotornis  purpuroptera,  Eiipp. 

Syst  Uebers.  No.  251,  p.  64,  t.  xxv. — HarUaub,  Joum.  £  Om.  1859, 

p.  11. — HeugL  Joum.  f.  Om.  1860,  p.  2. 
L.  anuus,  Brehm,  Habesch,  No.  82. 

Iris  yellow  white ;  legs  and  bill  black. 

Only  seen  in  the  Anseba  valley,  where  it  was  abundant, 
and  often  observed  associating  with  Lamprocolius  chaly- 
hems  and  X.  chrysogaster^  especially  about  villages  and 
cattle   enclosures.      It  was  occasionally  seen   hawking 


398  ZOOLOGY. 

insects  in  the  air,  as  was  also  L.  chrysogasler.  It  appears, 
however,  to  be  also  frugivorous  at  times.  More  than  three 
or  four  were  seldom  seen  together.  Dimensions  of  three 
specimens  : — 

Wing.        Tail.         BOL       Tanraa.  ''^^^iJ^t**'' 
in.  in.  in.  in.  in. 

1.  Male 676        86        076        16  14*0 

2.  „        6  7  8-2        0-8  1*6  13*8 

3.  Female     ....      6*66        8'3        076        1*6  14-0 

L.  cenea  has  black  spots  on  the  wing-coverts,  and 
the  back  greenish.  This  bird  has  unspotted  wing-coverts 
and  a  purple  beak,  and  the  measurements  are  different. 
The  specimens  obtained  greatly  exceed  Ruppell's  types 
of  L.  purpuroptera  in  size,  the  bill,  wings,  and  tail 
being  all  much  larger  than  the  specimens  in  the  British 
Museum.  But  the  coloration  is  identical,  while  that 
of  L.  cenea  is  quite  distinct 

xgo.  Amydrus  Riippelli,  Yerreauz. 

J.  Verr.  Chen.  EncjcL  d'Hist.  Nat  Ois,  t.  p.  166.— Hartlanb, 
Joorn.  £  Om.  1869,  p.  31.— Heugl  Joom.  f.  Om.  1869,  p.  12. 

Lamprotomis  morio,  Biipp.  Neo.  Wirb.  p.  26 ;  Syst.  Ueben* 
No.  262. 

Iris  red ;  bill  and  legs  black. 

This  race  abounded  around  Senaf^  and  elsewhere  in 
Tigrd  I  did  not  notice  any  Amydri  in  Lasta.  As  a 
rule  these  birds  kept  to  the  highlands,  at  about  from 
7,000  to  8,000  feet,  but  I  shot  one  specimen  in  May  as 
low  as  Suru,  barely  2,000  feet  above  the  sea.  The  fol- 
lowing measurements  are  taken  from  four  specimens 
of  each  sex  : — 

Wing  Toil.  Bill 

Male    ....  6-5  6*26  to  6*6        I'l  to  1-15 

Female      .     .     .       612  to  626  626  1'  to  1*06 


AFES.  399 

All  the  three  species  of  Amydrus  collected  by  me 
appeared  to  be  mainly  if  not  entirely  frugivorous,  living 
chiefly  on  the  fruits  of  various  kinds  of  FicuSy  of  Juni- 
pents  procera,  &c.  They  also  occasionally  feed  on  the 
ground  on  seeds.  They  roost  at  night  amongst  rocks 
in  large  communities. 

xgi.  A.  Blythii,  Hartl. 

Hartlaub,    Joum.    f.  Orn.   1859,    p.  32. — Heugl  Jonm.  f.  Om, 

1869,  p.  14. 
A.  RwppeUi,  Blyth,  Joom.  Ab.  Soc.  Beng.  1836,  p.  300. 

This  appears  to  be  a  fairly  distinguishable  race  so  far 
as  my  specimens  show.  It  has,  I  believe,  never  been 
brought  to  Europe  before,  the  only  known  examples 
being  in  the  Museum  of  the  Asiatic  Society  of  Bengal 
at  Calcutta,  brought  by  Captain  Speke  from  Somali 
land.  The  female  difiers  more  conspicuously  from 
A.  RUppelli  than  the  male  does ;  for  in  A.  Blythii  the 
whole  head  and  neck  are  ashy  grey,  the  colour  terminating 
abruptly,  and  not  passing  into  the  black  of  the  neck  and 
breast,  whilst  in  A.  RUppelli  the  feathers  have  black 
centres,  which  increase  in  size  below  until  all  become 
black.  Both  sexes  of  A.  Blythii  are  larger,  but  with  a 
smaller  bill,  than  A.  RUppelli.  In  the  former  species 
also  the  dusky  tips  to  the  chestnut  primary  quills  are 
larger  and  more  marked,  on  the  first  primary  especially ; 
nearly  two  inches  of  the  inner  web  and  from  two  and  a 
half  to  three  inches  of  the  outer  web  are  entirely  dusky  j 
whUst  in  A.  RUppelli  all  near  the  centre  is  rufous. 

As  this  race  appears  never  to  have  been  described 


400  ZOOLOGY. 

from  an  actual  specimen,  for  even  Heuglin  only  saw 
it,  and  failed  to  secure  examples,  I  add  the  following : — 
A.  Blythiiy  HartL — A.  similis  A.  Ruppelli  sed  mcyor^ 
rostra  brevioriy  colore  fusco  ad  apices  remigum  Tnagis 
extenso  capite  colloque  fcemincB  pure  ciiiereis^  hand 
nigro  strigatis. 

Long.  al».  Caudn.  Rostri  a  fr.    TanL     ^^^ 

in.  in.  in.  in.         inu 

Maris  dimensiones .     7*0  ad  7 '25    7*5   ad  8*25    0*9  ad  1*0    1*55     15*5 
FoBmina       „    .    .     6*37  „  6*75    6*67  „   7*25    0*9  „  0*95  1*5       14'5 

Amydrus  Blythii  abounded  around  Mayen  at  an 
elevation  of  from  3,000  to  4,000  feet  in  the  pass  below 
Senaf6.  In  January  and  February  they  were  frequently 
met  with  in  flocks,  which  were  often  seen  flying  around 
the  rocks,  their  red  wings  flashing  conspicuously  in  the 
sun.  I  once  found  a  great  roosting-place,  where  evidently 
large  numbers  from  all  quarters  collected  at  night 
It  was  in  a  ravine,  where,  owing  to  the  presence  of  a 
band  of  hard  rock,  the  watercourse  made  a  sudden 
descent,  and  here  even  in  the  dryest  season  a  very  small 
quantity  of  water  trickled  over  the  rocks.  As  in  many 
similar  places  where  water  trickled  down  the  face  of  a 
steep  cliff",  a  large  deposit  of  calcareous  tufa  had  taken 
place,  and,  probably  owing  to  the  subsequent  wearing 
action  of  the  torrent,  which  in  rainy  weather  rushes 
down  the  ravine,  had  been  so  excavated  as  to  form  an 
overhanging  cliff*,  beneath  which  these  birds  took  up 
their  abode  at  nightfall  in  great  numbers,  clinging  to 
the  rock.  The  water  also  attracted  numerous  Cynoce- 
phalus  hamadryas  Monkeys,  and  a  few  Pigeoi^s  and 
other  birds ;  the  principal   inhabitants,  however,  being 


AFJSS.  4U1 

the  Monkeys  and  Amydriy  and  the  noise  produced  by 
the  two  was  deafening,  especially  whenever  a  shot  was 
fired. 

192.  A.  albirostris  (Riipp.). 

PtiUmorhynchtu  alhirottris,  Eiipp.  NetL  Wirb.  p.  22,  pL  10,  f.  l,  2  ; 

Sjst.  Uebers.  No.  244. 
Ptilorhinus  albirostris,  Cab.  Mus.  Hein.  L  p.  201. — HartL  Jonm. 

f.  Om.  1859,  p.  30.— HeugL  Joum.  f.  Om.  1869,  p.  12. 

Iris  deep  coppery  red,  a  darker  ring  round  the  pupil  ; 
bill  white  ;  legs  black. 

To  separate  this  bird  genericaUy  from  the  last,  as  is  done 
by  many  ornithologists,  on  account  of  its  plumed  nares, 
is  simply  absurd,  when  in  every  other  detail  of  plumage, 
even  to  the  minutiae  of  colouring  in  both  sexes,  in  flight, 
and  in  habits  the  two  are  so  closely  allied.  Precisely 
like  A.  Ruppelliy  the  present  bird  is  social,  lives  on 
fruits,  and  roosts  in  large  communities  amongst  rocks. 
It  was  only  seen  on  the  highlands,  and  was  particularly 
common  in  the  neighbourhood  of  Senafd 

193.  Buphaga  erythrorhyncha  (Stanley). 

Taruigra  erythrorhyncha,  Stanley,  Appendix  to  Salt's  Journey,  App. 

p.  lix. 
B,  habessiniea,  Hemp,  and  Ehr.,  Symb.  Phys.  t.  ix. 
B.  erythrorhyncha,  Riipp.  Syst.  Uebeis.  No.  254. 

Iris  scarlet,  orbit  yellow;  bill  within  and  without, 
and  tongue,  bright  scarlet,  dusky  in  the  young;  feet 
dusky. 

The  habits  of  this  bird  are  well  known.  It  usually 
abounds  wherever  cattle  are  to  be  met  with,  and  perches 
on  oxen,  mules,  or  camels,  running  over  their  backs,  and 

D  D 


402  ZOOLOGY. 

up  and  down  their  sides,  seaxching  for  parasites,  especially 
Ixodes,  on  which  it  feeds,  and  frequently  appearing  to 
defy  the  attempts  of  the  animal  to  induce  it  to  leave. 
The  number  of  ticks  which  assail  both  wild  and  domestic 
animals  in  Abyssinia,  especially  in  the  rainy  season,  is 
astonishing;  not  even  the  thick  hard  hide  of  the  Rhino- 
ceros is  proof  against  them,  though  the  species  found 
upon  him  appears  to  be  peculiar.  I  did  not,  how- 
ever, notice  this  bird  on  a  Rhinoceros,  or  any  other 
wild  animal 

Buphaga  erythrorhyncha  is  found  both  on  the  high- 
lands and  near  the  coast.  It  is  more  abundant  in  the 
former.  The  noise  is  much  like  that  of  a  Starling 
or  Myna. 

194.  Textor  alecto,  Temm. 

Temminok,  PL  CoL  446.— Eiipp.  Syst  Ueben.  No.  257.— Hengl. 
Joorn.  f.  Orn.  1867,  p.  366. 

Legs  dusky,  horn-coloured ;  bill  black,  the  basal  portion 
in  adults  (males)  only  covered  with  a  fleshy  cere,  pale 
yellow  in  colour. 

I  saw  this  bird  only  on  the  Anseba.  It  is  quite 
starling-like  in  its  habits  and  flight,  and  belongs,  I  think, 
to  this  family  rather  to  the  FringiUidcB.  I  frequently 
saw  it  associating  with  Lamprotomis  mneus,  Lampro- 
colius  chalyhcBtiSy  and  X.  chrysogastevy  hunting  for  insects 
on  the  ground,  especially  about  cattle-pens.  The  massive 
nests  are  not  unlike  those  of  Stumopastor  contra.  All 
which  I  saw  were  in  high  acacia-trees,  but  the  breeding 
season  was  over  long  before  July. 


AFES.  4()a 

Family  FRINGILLIDiE. 

195.  Hyphantornis  larvata,  Eupp. 

Biipp.  NeiL  Wirb.  p.  91,  t.  xxxiL  f.  1 ;  Syst  Uebers.  Nx).  260. — 

HeugL  Journ.  f.  Om.  1867,  p.  378. 
F^flavaviridis,  Riipp.  Syst.  Uebers.  p.  69,  t  xxix. 

Iris  scarlet ;  beak  black ;  legs  greyish  black. 

Only  met  with  about  Dolo  and  Agula  in  Tigrd, 
about  half-way  on  the  road  between  Adigrat  and 
Antalo;  at  these  places  it  abounded.  Scarcely  any 
birds  had  attained  their  full  breeding  plumage  in  the 
middle  of  May. 

The  eggs  are  stated  by  Von  Heuglin  to  be  pale  bluish 
green,  with  a  few  violet-brownish  spots  at  the  blunter 
end. 

Ploceus  JlavoviridiSy  Riipp.,  is  the  same  bird  iA  winter 
plumage,  when  the  bill  is  horn-coloured. 

196.  Hyphantornis  melanotis  (Gu^r.). 

PlocexLs  vielanotis,  Gu6r.  Rev.  ZooL  1843,  p.  321  (nee  LafresnAye).— 

Lefebvre,  p.  110,  Atlas,  pL  ix. 
P.  Gvhiniy  G.  R.  Gray,  Gen.  of  Birds,  ii.  p.  351. — Ferr.  et  GtaL 

No.  110.— Heugl.  Joum.  f.  Om.  1867,  p.  388. 

Iris  yellowish  grey ;  beak  black  in  the  male,  horn- 
coloured  in  both  sexes  in  non-breeding  plumage,  and  in 
the  female  always;  legs  flesh  colour. 

Common  throughout  the  highlands,  and  seen  at 
nearly  10,000  feet,  though  in  general  more  common 
at  a  rather  lower  elevation.  Specimens  shot  at  Senaf^ 
jit  the  end  of  May  had  assumed  the  full  breeding 
plumage. 


404  ZOOLOGY. 

297.  H.  galbula  (RtLpp.). 

Tlouva  gaOndtUy  Riipp.  Nen.  Wirb.  p.  92,  t  xxxiL  I  2 ;  Syvt 
Uebera.  No.  261.— Ferr.  et  GaL  No.  109.— Lefebyre,  p. 
109.— Brehm^  Habesch,  No.  87. — Hengl  Jouzn.  t  Orn.  1867, 
p.  385. 

Iris  orange  brown  ;  bill  black — in  the  male  in  breeding 
plumage  only  it  is  horn-coloured,  the  upper  mandible 
being  darker  in  the  female,  and  in  the  male  in  non- 
breeding  plumage ;  legs  flesh-coloured. 

Abundant  on  the  shores  of  Annesley  Bay,  in  the  passes 
up  to  about  4,000  feet,  and  in  the  Anseba  valley.  In 
the  latter  locality  it  breeds  in  August;  at  the  com- 
mencement of  the  month  the  nests  were  all  built, 
but  no  eggs  were  laid.  The  nest  is  of  gra^s.  as  usual 
shaped  like  a  pear,  with  an  opening  at  the  side  without 
any  long  tubular  entrance,  and  is  often  hung  on  acacia 
trees. 


198.  H.  luteola  (Licht.). 

FringiUa  luUoloy  Licht.  Yen.  d.  DoabL  p.  23,  nota. 
FlociUi  perionaJtuSf  VieilL  Gal.  pL  Ixxxiv.  p.  117. 
HyphanUyrnis  chrywmelaSy  Heagl  Jouid.  £  Orn.  1862,  p.  25. 
Hyphantom%8  penonata,  Heugl.  Joam.  £  Orn.  1867,  p.  390. 
H.  luUolOy  Finsch,  Jouzn.  £  Orn.  1868,  p.  170. 

Iris  burnt  sienna  ;  bill  black  (in  male  only) ;  legs  horn 
colour. 

Rare  in  the  Anseba,  and  not  seen  elsewhere.  It  was 
breeding  in  the  commencement  of  August.  Mr.  Jesse 
obtained  the  nest,  which  is,  as  usual,  bottle-shaped,  but 
with  a  long  tubular  entrance.  It  contained  two  white 
eggs. 


V 


AFES.  405 

199.  Euplectes  xanthomelas,  Biipp. 

Eiipp.  Nea.  Wirb.  p.  94  ;  Syst.  Uebers.  No.  266,  p.  67,  t  xxTili. — 
HeugL  Joum.  f.  Om.  xy.  1867,  p.  375. 

This  handsome  species  ranged  to  a  greater  height  above 
the  sea  than  any  other  Weaver-bird  met  with.  It  was 
common  at  10,500  feet  on  the  Wandaj  pass  and  at 
Santara  in  Dalanta.  I  also  found  it  at  Ashangi  at 
8,000  feet,  but  never  noticed  it  in  any  other  locality.  It 
had  not  quite  assumed  the  fuU  plumage  at  the  beginning 
of  May. 

200.  Quelea  sethiopipa  (Sand.). 

Floceus  athiopicus,  Sundeyall,  Ofr,  k.  Vet  Akad.  Forh.  1850, 

p.  126. 
Euplectes  quelea  ou  tanguinirostris,  Ferr.  et  GraL  No.  112. 
Q.  tanguinirostrU  orierUcUis,  Heugl  Joum.  f.  Om.  xv.   1867, 

p.  391. 

Bill  in  male  deep  blood-red,  paler  towards  the  tip ; 
orbit  orange  red ;  legs  brown.  In  the  female  both  bill 
and  legs  are  much  paler. 

Only  met  with  in  Northerji  Abyssinia ;  it  was  first  seen 
at  Ailat,  and  occurred  abundantly  in  the  Anseba  valley. 
It  was  several  times  seen  feeding  on  insects,  and  even 
pursuing  winged  ants  and  catching  them  in  the  air.  It 
was  always  in  flocks,  and  did  not  appear  to  be  breeding 
in  the  rains. 

201.  Coliuspasser  laticauda  (Licht.).. 

FrififfiUa  laticauda^  Licht  Verz.  der  DoubL  p.  ^4. 
Coliuspasser  torqiuitus,  Riipp.  Neu.  Wirb.  p.  36. 
Vidua  latica/uda,  Heu^  Syst  Uebeis.  No.    273.— Joom.  t  Om. 
1867,  p.  .396. 

Only  met  with  on  two  occasions  on  the  highlands, 
once   near  Antalo,  the  other  time  near  Agula,  both  in 


406  ZOOLOGY, 


May.     It  was  in  flocks  in  both  instances,  and  very  wary. 
It  feeds  on  the  ground. 

A  pair  shot  in  the  beginning  of  May  measure  : — 


iiri««        T.n      T.«ni.    Billflrom  Whole  length 
Wing.       Tail.     Tarsus.  fo„head.        about 

in.  in.  in.         in.  in. 


Male 
Female 


3-3 

5-6 

1-0 

0-6 

9-0 

3-2 

4-3 

1-0 

0-56 

76 

All  the  specimens  killed  by  me  were  in  winter  plumage, 
although  they  had  assumed  the  long  tail  The  whole 
upper  surface  in  both  sexes  is  black,  with  pale  brown 
margins  to  the  feathers  of  the  head,  neck,  and  intersca- 
pularies — much  less  marked,  however,  on  the  last — and 
whitish  on  the  rump.  Wings  and  tail  black,  secondaries 
and  outer  rectrices  with  brown  margins.  Lower  surface 
pale  isabelline  with  black  striae,  yellowish  on  chin  and 
throat  in  some  specimens.  There  is  a  distinct  pale 
supercilium. 

On  closely  examining  the  feathers,  it  is  evident  that  a 
change  of  colour  is  taking  place,  on  the  head  especially ; 
here  and  there  portions  of  the  web  are  becoming  reddish  ; 
and  it  is  evident  that  in  this  bird  the  change  from  black 
and  brown  to  red  takes  place  without  a  moult 

I  cannot  help  feeling  surprised  that  so  few  naturalists 
admit  Coliuspasser  as  a  distinct  genus  from  Vidua. 
Both  have,  it  is  true,  lengthened  tails ;  but  there  their 
resemblance  ends.  In  Vidiia  only  the  central  tail-feathers 
are  elongated,  and  they  are  greatly  changed  in  form  ; 
in  Coliuspasser  all  are  lengthened,  and  no  modification 
takes  place  in  form.  Vidua  has  the  bill  of  the  Estrildince^ 
Coliuspasser  that  of  the  PloceincB, 


AFES.  407 

202.  C.  flavi-scapulatus>  Biipp. 

Eiipp.  Neu.  Wirb.  p.  98. 

Vidua  macroeerca,  HeugL  Joam.  t  Om.  1867,  p.  394. 

I  am  indebted  for  the  only  specimen  in  my  collection 
to  Mr.  Jesse,  who  shot  it  at  Guna  Guna.  It  is  in  winter 
dress.  The  upper  part  of  the  wing  above  being  brilliant 
yellow,  and  the  rectrices  not  produced  in  this  dress,  it 
has  all  the  appearance  of  an  Euplectes. 

In  this  dress  it  may  be  distinguished  from  the  southern 
C.  macrocerca,  Licht.,  by  the  tips  of  the  qmll  feathers 
being  dark.  In  C.  irmcrocerca  they  are  pale.  This  was 
pointed  out  to  me  by  Dr.  Finsch.  In  full  plumage  the 
two  birds  arc  very  distinct,  the  interscapularies  being 
yellow. 

203.  Vidua  sphenura,  Verr. 

Bp.  Consp.  Qtea,  Ay.  L  p.  449. 

V,  paradisea,  Eiipp.  Sjst  Uebers.  No.  270. — ^Brehm,  Habesch, 

No.  217. 
V,  sphenuray  Heugl.  Joom.  f.  Om.  1867,  p.  393, 

Not  rare  at  Komayli,  at  the  base  of  the  hills  west  of 
Zulla,  in  February  and  March,  but  not  seen  anywhere 
else ;  and  it  had  apparently  disappeared  from  the  dis- 
trict in  May.  The  male  in  full  plumage  is  equal  to 
V.  paradisea,  L.,  in  size.  The  following  are  the  dimen- 
sions of  a  fine  male  skin  for  which  I  am  indebted  to 
Captain  Sturt.  The  length  exceeds  that  given  by  Von 
Heuglin : — 

in. 

Total  length 1676 

TaU 12-5 

Outer  tail-feathers 2*3 

Wing 3-2 

Beak 64 

Tarsus 0*7 


an.  Pjrtelia  citerior,  StrickL 

Stzickland,  CoDtribntioiu  to  OniitlL  1853.  p.  IM. 
Eririlda  dtgan*,  Bupp.  S71L  Ueben.  Na  S7& 
ZtmagaitTu  ciierior,  HengL  Joam.  t  Om.  1868,  p.  19. 

Iris  sieiiDa  brown ;  beak  light  red  ;  legs  flesh  colour. 

Occaflionally  met  with  near  the  coast,  and  also  in  the 
Anseba  valley  up  to  4,000  or  4,500  feet  above  the  sea. 

The  bill  in  my  specimen  measures  nearly  half  an 
inch.    Von  Heugtin  gives  4^  Paris  lines. 

313.  Emberiza  septemstriata,  Biipp. 

Lefeb.  p.  120.— Brehm,  Habesch,  No.  96.— Rfipp.  Nen.  Wiife.  p.  86, 
t  jxx.  t  i ;  Sj>t.  XJebeM.  Nft  899.— HengL  Jonra.  £  Orn. 
1868,  p.  77. 

Iris  dark  brown;  upper  mandible  dusky,  lower  yel- 
lowish ;  feet  horn  colour. 

Abundant  at  places  where  water  occurred  near  the 
coast,  as  at  Saati,  Ailat,  &c.  in  Samhar,  in  July  and  August. 
Found  also  at  Ain,  and  more  rarely  at  higher  levels.  I 
shot  one  specimen  at  Senaf<^  at  the  end  of  May. 

Von  Heuglin  is  probably  correct  in  considering  this 
Bunting  as  a  permanent  inhabitant  It  very  likely 
leaves  the  coast  plain,  however,  in  winter,  like  many 
other  birds. 

In  KUppell's  description  of  this  species  the  sexes  are 
said  to  be  similar,  and  the  head  is  described  as  black 
above  with  a  white  streak  in  the  centre.  In  the  speci- 
mens collected  by  me  the  male  alone  has  the  head  black 
above,  and  the  central  stripe  above  is  grey,  not  white, 
the  three  side  stripes,  through  the  eye  and  above  and 
below  it,  being  white ;  the  thin  is  white  with  a  central 


AFE8.  409 

Anseba.    The  dimensions  of  my  specimens  slightly  ex- 
ceed those  given  by  Von  Heuglin  : — 


Female 


wing, 
in. 

TaiL 
in. 

Tanus. 
in. 

BiU. 
in. 

1-9 

1-86 

0-6 

0-36 

1-9 

176 

0-6 

0-34 

208.  E.  minima,  YieilL 

Vieillot,  Ois.  Chant,  t  x. — Biipp.  Syst  Uebera.  No.  276. — HeogL 
Joum.  t  Om.  1868,  p.  14. 

Iris  brown,  orbit  yellow  ;  bill  crimson,  except  a  dusky 
line  along  the  commissure,  pink  near  the  base;  legs 
brown. 

Not  unfrequently  met  with  on  the  highlands,  and 
also  in  the  Anseba  valley. 

209.  E.  quartinia,  Bp. 

Bp.  Consp.  Gen.  Av.  i.  p.  461. 

E.  Emesti  IL  HengL  Joum.  f.  Om.  1862,  p.  29,  et  1868,  p.  10. 

Iris  crimson ;  legs  dusky ;  upper  mandible  black,  lower 
crimson. 

Two  specimens,  both  females,  were  obtained  at  SenaK. 
One   of  them   measures:  wing  1*8  in.,  tail  1*6,  tarsus 

■ 

0*6,  bill  from  forehead  0*3  in. 

210.  E.  phcenicotis,  Swains. 

Swainson,  Birds  of  W.  Afr.  p.  192,  pi.  xiv. — Heogl.  Joum.  L  Om. 

1868,  p.  18. 
E,  btTigcUus,  Riipp.  Syst.  Uebers.  No.  275. 

Common  on  the  highlands,  less  so  at  a  lower  elevation, 
but  seen  not  infrequently  in  the  Anseba  valley,  and  even 
as  low  a^  Ailat  (1,200  feet  above  the  sea)  in  Samhar. 


211.  Pytelia  citerior,  StrickL 

Strickluid,  Contribattoiu  to  Omith.  ISAi,  p.  160: 
Ei^Uda  dtgrau,  Biipp.  Sy»t  Uobere.  No,  278, 
Zonagiulrit  eittrior,  HengL  Joiim.  f,  Otd.  1868,  p.  19, 

Iris  sienna  brown ;  beak  ligbt  red  ;  legs  flesh  colour. 

Occasiooally  met  with  near  the  coast,  and  also  in  the 
Anseba  valley  up  to  4,000  or  4,500  feet  above  the  sea. 

The  bill  in  mj  specimen  measures  nearly  half  an 
inch.     Von  Heuglin  gives  4^  Paris  lines. 

313.  Emberiza  septemstriata,  RUpp, 

Lefeb.  p,  120,— Brebm,  Habesch,  No.  96.— Biipp.  Nen.  Wirb.  pL  86. 
t  XIX.  t  2 ;  Syst.  Uebere.  Na  299.— HengL  Jonm.  £  Om. 
1868,  p,  77. 

Iris  dark  brown ;  upper  mandible  dusl^,  lower  yel- 
lowish ;  feet  horn  colour. 

Abundant  at  places  where  water  occurred  near  the 
coast,  as  at  Saati,  Ailat,  &c.  in  Samhar,  in  July  and  Augusts 
Found  also  at  Ain,  and  more  rarely  at  higher  levels.  I 
shot  one  specimen  at  Senaf^  at  the  end  of  May. 

Von  Heuglin  is  probably  correct  in  considering  this 
Bunting  as  a  permanent  inhabitant.  It  very  likely 
leaves  the  coast  plain,  however,  in  winter,  like  many 
other  birds. 

In  BUppell's  description  of  this  species  the  sexes  are 
said  to  be  similar,  and  the  head  is  described  as  black 
above  with  a  white  streak  in  the  centre.  In  the  speci- 
mens collected  by  me  the  male  alone  has  the  head  black 
above,  and  the  central  stripe  above  is  grey,  not  white, 
the  three  side  stripes,  through  the  eye  and  above  and 
below  it,  being  white ;  the  fchin  is  white  with  a  central 


AVES,  411 

black  mark ;  throat  cinereous,  becoming  blackish  below. 
In  the  female  the  head-feathers  are  dusky  brown  with 
pale  margins,  the  central  stripe  above  being  obsolete  or 
rudimentary,  and  the  lateral  stripes  isabelline  :  chin  and 
throat  pale  brown  with  fine  black  marks.  This  is  E. 
capistrata,  Cab.,  to  judge  from  the  specimen  in  the 
Berlin  Museum. 

213.  E.  flaviventris,  Vieill. 

Vieillot,  Encyc.  Meth.  929. 

E.  flomgoMer,  Kiipp.  AUas,  t.  xxv;   Syst.    Uebers.   No.   298. — 
HeugL  Joum.  f.  Om.  1868,  p.  75. 

Bill  brown ;  upper  mandible  dusky ;  legs  brown. 

Not  a  common  bird  in  the  Anseba  valley,  and  met 
with  also  in  Habab  at  about  3,000  feet  above  the  sea. 
Captain  Sturt  shot  specimens  in  a  valley  below  Senafd 
Dimensions : — 


Wing, 
in. 

Tail. 
In. 

Tubus. 
in. 

BiU. 
in. 

315 

2-8 

0-7 

0-45 

31 

2-6 

07 

0-4 

Male 

Fenude 3*1 

2x4.  Passer  Swainsoni,  RUpp. 

Emberxza  capenns,  Salt,  No.  46,  App.  p.  xlyiii. 

Pyrgiia  Bwainsonii^  Riipp.  Neu.  Wirb.  p.  94,  t.  xxxiiL  f.  2. 

P.  simplex,  Svraiiis.  Birds  W.  Afr.  p.  208. 

P.  Sinainsonii,  Riipp.  Syst  Uebers.  No.  296. 

Iris  brown. 

The  common  Sparrow  of  the  Abyssinian  highlands, 
abundant  about  all  villages:  it  is  also  common  in 
woods,  away  from  human  habitation.  It  is  rare  at 
lower  elevations,  though  I  met  with  it  once  at  Ailat. 

In  RUppcU's  figure  the  white  bars  on  the  wing- 
coverts  are  omitted,  and  the  ferruginous  rump  is  neither 


412 


ZOOLOGY. 


1/ 


shown  in  the  drawing  nor  mentioned  in  the  description. 
The  dimensions  also  appear  too  smalL 

2x5.   P.  {Gymnorii)  canicapillus  (1),  Bljth. 

Blyih,  Ibis,  1865,  p.  46. 

P.  supra  hrunneus,  eapiU  ciiurto^  iuperciliii  latii  cinnamaneiapaatice 
in  laterUms  cervicU  diffutis :  rectricibus  remdgibuiqtufuligenanSf 
paUido-marginatis ;  genisy  UUeribus  colli  et  pectore  sordide 
eanis,  mento,  gula  ahdomineque  medio  aJhe&cenHlm$  macula 
gvlcurifiavo  hand  conspicuo :  iridibus  olivaceis,  pedihus  rubella- 
eomeis,  mandibuld  fused  maxUld  palide  eomed. 


Long.  iota. 

AL 
in. 

Caad. 
in. 

Tani. 
in. 

Bottr.  afr. 
in. 

1.  Foemina 

.     .     .       6-0 

31 

2-0 

0-66 

0-45 

2.        „ 

.     .     .       4-76 

315 

1-9 

0-65 

0-6 

3.  Mas? 

.     .     .       476 

30 

1-8 

0-65 

0-4 

I  found  this  bird  solitary  or  in  small  flocks  both  in 
the  woods  and  around  villages  in  the  Lebka  valley  and 
its  neighbourhood,  at  3,000  to  4,000  feet  above  the  sea. 
It  was  especially  abundant  at  the  village  of  Rairo  in 
Habab. 

It  is  distinguished  from  Xanthodina  dentata^  Sunde- 
vall  ("Ofvers.  k.  Vetensk.  Acad."  Stock.  1850,  p.  127), 
and  Petronia  alhigularisy  Brehm  ('^  Naumannia,"  1856, 
p.  377),  which  Dr.  Finsch  considers  to  be  the  young  of 
X.  dentata,  by  having  the  eyebrow  cinnamon  instead  of 
pale  rufous,  by  the  head  being  more  grey,  and  by  the 
back  being  browner  and  more  uniform  in  coloration. 
Mr.  Blyth  considers  that  it  is  his  canicapilluSy  and  it 
only  differs  from  the  description  in  being  a  little  smaller 
than  Passer  {Gfymnoris)  Jlavicollis,  FrankL,  while  Bly th's 
type  was  of  the  same  size.  The  specimen  in  the  Liver- 
pool Museum  described  by  Blyth  was  probably  from 
South  Africa. 


A  FES.  413 

2i6.  P.  (Gymnoris)  pyrgita  (Heugl.). 

Xanthodina  pyrgita^  HengL    Jonm.  f.  Om.   1862,  p.  30 ;  ibicL 
1868,  p.  80. 

Only  killed  once.  A  female  was  shot  in  the  Komayli 
pass  near  Suru.  It  is  a  much  larger  bird  than  the  last, 
with  far  more  uniform  colouring.  Dimensions :  wing 
3*3  in.,  tail  2*5,  tarsus  075,  bill  from  forehead  0'45, 
whole  length  about  6  in. 

217.  Fringilla  (Polyosptza)  tristriata  (Riipp.)- 

Serinus  tristriatus,   Riipp.   Neu.  Wirb.  p.   97,  t.   xxxv.  f.   2.— 

Lefebvre,  p.  118. — Riipp.  Syst  Uebers.  No.  289. 
F,  tristriaU,  HeugL  Jonm.  f.  Om.  1868,  p.  91. 

Iris  brown,  pupil  small. 

Very  common  throughout  the  highlands,  generally 
amongst  bushes,  in  small  flocks  or  singly.  It  keeps 
much  to  the  ground.     I  never  saw  it  at  lower  elevations. 

218.  F.  (Foliotpiza)  xanthopygia  (Riipp.). 

Serinus   xanthopygivSy  Riipp.    Neu.   Wirb.  p.  96,  t.  zxxv.  f.  1 ; 
Syst.  Uebere.  No.  288.— Heugl  Joum.  f.  Om.  1868,  p.  90. 

I  only  twice  obtained  specimens  of  this  species,  both 
on  the  highlands,  one  near  Senaf^,  the  other  south  of 
Antalo.     Mr.  Jesse  shot  it  in  the  Anseba  valley. 

219.  Crithagra  striolata  (Riipp.). 

Fyrrhula    striolata^  Riipp.    Neu.  Wirb.  p.  99,  t.  xxxxiL    f.  1  ; 

Syst  Uebere.  No.  317. 
Serinus  striolaius,  Heugl.  Journ.  f.  Om.  1868»  p.  94. 

Iris  brown. 

This  is  not  a  veiy  common  bird,  and  it  was  only 
observed  on  the  higher  levels,  usually  solitary,  but  occa- 
sionally in  flocks.     I  met  with  it  above  9,000  feet. 


4U  ZOOLOGY. 

220.  Crithagra  flavivertex,  W.  Blanford  (Plate  VIL), 

C.  mpra  olivcuiea,  obsolete  fusco^mouiulata  fronU  laid  aured,  colore 
ad  9ummo  pileo  paulatim  olivcueeniiy  mpereiliis  flavxSj  postiee 
productis,  loris  fuscisy  colli  laicrihus  oliva^ceis,  imnuwuleUUy 
uropygio  Jlavo,  remigibus  cum  teetridlms  alarum  rectrieibtuqiu 
fuscMy  flavo-marginaiis,  remigium  Tnarginibus  intemU  pctUidit 
gastrcso  sordideflavo,  olivascmti-lavato,  erxsso  olbeaeenH  ;  rostra 
hruneoj  pedibus  fusois, 

Fcemina  vix  dHutius  eolorata, 

^'  Loiig,  dUx  maris  3*15,  tarsi  0*6,  rostr,  a  fronU  0*35  ;  /oBtnttMB,  al, 

3*05,  Cauda  2*25  :  long,  tola  circa  5*25  poll.  Angl  et  dee. 

This  bird  is  closely  allied  to  Cntliogra  htUpxtcea, 
L.,  and  C.  canioollisy  Swains.  It  may  be  distinguished 
from  the  first  by  its  yellow  forehead  being  broader  and  by 
the  colour  passing  gradually  to  olivaceous  on  the  crown, 
whereas  in  C.  hutyracea  the  yellow  is  a  narrow  band 
sharply  defined.  There  are  several  minor  differences  also. 
From  C.  canicollis  it  is  distinguished  by  the  back  and 
sides  of  the  neck  being  olive  instead  of  bluish  grey,  and 
by  both  the  forehead  and  underparts  being  brighter  in 
colour.  It  is  a  much  greener  bird  than  Serimis  aurifrons, 
Tristram,  larger  in  size  and  with  a  less  conical  bilL  C. 
chloropsisy  Cab.,  from  Mozambique  is  distinguished  by 
having  a  much  larger  bill  than  the  present  species,  larger 
wings,  and  less  yellow  coloration. 


1  f 

I 


r 


i 

;    t 


* 


■ 


221.  Serinus  citrinelloides,  Kiipp. 

Riipp.  Neu.  Wirb.  p.  95,  t.  xxxiv.  f.  1 ;  Syst  Uebera.  No.  286. 
Fringilla  citrindloides,  UeugL  Joum.  f.  Om,  1868,  p.  92. 

Iris  brown. 

This  bird  agrees  very  poorly  with  Ruppell  s  description, 
but  better  with  his  figure,  and  still  better  with  Von 
Ht'Uglin's  description.  I  shot  a  pair  at  Senaf(5  on  a  large  fig- 


fN 


CRITHAGRA    FLAVIVERTEX.  W  Slinf 


I 


I 

I 


•I 


; 


AFES.  415 

tree.     The  dimensions  are  :  wing  2*65  in.,  tail  1*9,  tarsus 
0*54,  bill  from  forehead  0*37,  whole  length  about  4*25  in. 

222.  S.  nigriceps,  Kiipp. 

Riipp.  Neu.  Wirb.  p.  96,  t.  xxxiv.  I  2  ;  Syst.  Uebera.  No.  287.— 

Lefebyre,  p.  118. 
FringiUa  nigriceps,  HeogL  Journ.  t  Orn,  1868,  p.  93. 

Only  seen  at  considerable  elevations,  but  very  common 
on  the  passes  above  9,000  feet.  It  appears  to  belong 
to  the  Subalpine  fauna,  though  a  specimen  was  shot 
by  my  collector  at  Adigrat  It  was  always  met  with 
in  flocks,  keeping  to  open  ground,  and  often  seen  on 
ploughed  fields. 

Order  OEMITOBES. 
Family    COLUMBIDiE. 

223.  Columba  guinea,  L. 

LiniUBUs,  Syst  Nat  ecL  12»,  p.  282.— Salt,  No.  47,  App.  p.  xlviil 
—Kiipp.  Syst  Uebera.  No.  363. 

Iris  bronze  yellow  within  near  the  pupil,  copper 
colour  extemaUy. 

This  bird  lives  about  rocks,  also  on  Kolqual  trees^  around 
churches,  and  probably  occasionally  about  houses.  It  is 
often  seen  in  flocks  in  the  fields.  In  the  passes  below 
Senaf^  it  was  not  observed  below  4,000  feet,  but  in  the 
Lebka  it  occurred  at  only  2,000  feet  above  the  sea. 

I  should  scarcely  call  it  a  domestic  Pigeon,  as  some 
writers  do.  It  certainly  is  not  nearly  so  constantly  seen 
about  villages  as  C.  intermedia  is  in  India,  though  it 
is  often  met  with  in  the  iieighl>ourhoo<l  of  houses. 


416  zooLoar. 

334.  C.  albitorques,  Rilpp. 

Riippell,  N«iL  Wirb.  p.  63,  t  xxii.  f.  I.— Syst.  Ueben.  No.  364. 

Iris  dull  red,  approaching  violet ;  beak  black,  cen 
white ;  legs  pink. 

This  bird  is  somewhat  intcnnediate  in  its  cbaracten 
between  Columba  and  Turtur,  but  in  its  habits  it  ii 
rather  a  Pigeon  than  a  Dove,  keeping  in  flocks,  and 
usually  on  the  ground.  It  is  also  excellent  eating 
not  bitter  and  tough  like  most  Doves.  It  was  onlj 
seen  on  the  highlands,  in  the  temperate  region. 

335.  Turtur  semitorquatus  (Kiipp.), 

Colum&a  ttmiUnripuaa,  Riipp.  Neo.  Wirb.  p.  66,  pL  83,  fig.  2. 
Turtur  erytAnipAryi,  Swum.  Bink  W.  A&.  ii  p.  807,  (^  SS. 
Turtur  Mmitor^uotiM,  Riipp.  ajst.  Ueben.  No.  369. 

Iria  orange,  orbit  puq)lish ;  red  legs ;  as  also  the  bill 
black. 

I  first  met  with  this  large  Dove  about  Lake 
Ashangi,  and  in  the  valleys  further  south  about  Lat 
and  Dildl  It  was  not  seen  on  the  highlands  of  Tigr^, 
but  it  occurred  abundantly  at  Ain,  on  the  Lebka.  in 
Habab  at  a  very  low  elevation,  and  in  the  Anseba 
vaUey.     It  always  keeps  to  thick  bushes  and  trees. 

326.  T.  lugens  (Riipp.). 

Colimha  Utgtni,  Riipp.  Nen.  Wirb.  p  64,  t.  zxil  C  % 
ISirtur  lugtiu,  Riipp.  SjrRt.  Ueben.  No.  368. 

Iris  orange  yellow,  with  a  tinge  of  orange ;  feet  pink ; 
bill  reddish  dusky. 

The  common  Dove  of  the  Abyssinian  highland&  I  did 
not  notice  it  below  6,000  feet. 


I  -'-FN. 


Arss.  417 

227.  T.  albiventris,  G.  R  Gray. 

Gray,  Gknera  of  Birds,  voL  iL  p.  472. 

T.  $emiiorquatus,  Swains.  Birds  W.  Afr.  iL  p.  108. 

T,  risoriusy  Riipp.  Syst  Uebers.  No.  366. 

Iris  crimson ;  beak  dull  slaty ;  legs  purplish  red. 

The  common  Dove  of  the  coast  country,  far  more 
abundant,  however,  after  the  rains  in  February  than 
before. 

228.  T.  senegalensis,  L. 

Riipp.  Syst  Uebers.  No.  367. 

Iris  dark  brown. 

This  Dove  ranges  in  altitude  to  a  greater  extent 
than  the  preceding  species.  It  is  most  abundant  in 
the  subtropical  belt,  from  3,000  to  6,000  feet,  but 
occurs  both  above  and  below  these  limits. 

229.  Peristera  afra  (L). 

Syst.  Nat  ed.  12»,  I  p.  284. 

P.  chaleopsUos,  Wagler,  Systema  Aviam,  No.  83. — Riipp.  Syst. 
Uebers.  p.  98,  t.  xxxviiL 

Iris  brown  ;  bill  deep  purple ;  legs  purplish  red. 

A  ground   Dove,  which  keeps   itself  almost   always 
amongst  bushes  and  thick  underwood.    It  was  seen  com- 
monly in  the  thick  jungles  along  the  banks  of  ravines 
and  watercourses  in  Samhar  and  Habab,  especially  about 
Ailat  and  Ain.     Near  Annesley  Bay  it  was  not  met 
with,   nor  in  the  passes  below   Senaf^,   except  about 
Mayen   and   Undul.      On  the  highlands  it  was  only 
observed    in  the  western  valleys,   about  Marawa  and 
Dildi,  south  of  Lake  Ashangi     It  is  a  shy,  silent  bird 

E  £ 


330.  MnBL  capensis  (L.). 

Colianba  eaptruit,  Linn.  Sjat.  Nkt  ed.  1S„  L  p.  S86L 
^na  capmtit,  Kiipp.  Sjat.  Ueben.  No.  371. 

Ilia  brown ;  basal  half  of  bill  purplish  red,  tip  yellow 
legs  pink.  In  yoting  birds  the  plumage  is  mottlec 
the  bill  dusky,  all  of  one  colour,  and  the  legs  purplis 
brown.  They  were  in  this  plumage,  and  evidently  ful 
grown,  in  August. 

I  first  met  with  this  little  Dove  close  to  Dongolc 
at  about  6,500  feet  above  the  sea,  halfway  betweei 
Adigrat  and  Antalo.  It  is,  however,  rare  on  the  high 
lands.  In  December,  January,  and  February,  there  wer 
none  whatever  near  Annesley  Bay,  but  after  the  rail 
in  the  latter  month  numbers  appeared,  and  in  May  ant 
June  it  was  one  of  the  commonest  birds  about  Zull 
and  Komayli.  It  also  abounded  in  Samhar  and  th 
tropical  portion  of  Habab,  becoming,  however,  very  rar 
or  wanting  in  the  subtropical  or  Anseba  'raUey. 


Fakilt  TRt3lONID£ 

331.  Treron  abyssinica  (Lath,). 

Wulik,  Bmoe,  Tnveli,  voL  v.  p.  166. 

Columfta  abytriniea,  LatL  lad.  Om.  Snpp.  p.  fac. — Salt,  No.  4 

App.  p.  xlriiL 
Vinogo  obytftntea,  Riipp.  Syit.  Ueben.  No.  360. 

Iris  yellow  or  salmon  colour,  with  a  circle  of  pm 
blue  round  the  pupil ;  bill  dirty  white ;  cere  purplis 
pink  :  legs  deep  yeUow. 

This  bird  is  very  closely  allied  in  its  form  and  habii 
to  the  Indian  Oreen  Pigeon 


AVES.  419 

and  its  allies,  and  its  call  is  a  very  similar  liquid  whistle, 
slightly  differing  in  tone,  and  with  the  concluding  por- 
tion a  little  harsher  and  more  prolonged,  but  it  was 
recognised  immediately  by  all  who  were  familiar  with 
the  peculiar  note  of  the  Indian  bird.  It  feeds  on  fruit, 
especially  those  of  different  kinds  of  Ficus. 

The  Abyssinian  Green  Pigeon  is  chiefly  found  in  the 
subtropical  belt.  I  very  rarely  saw  it  on  the  highlands. 
In  the  passes  below  Senaf^,  it  first  appeared  above  Mayen 
or  Undul  Wells,  at  about  4,000  feet  above  the  sea, 
and  it  was  not  seen  above  6,000  feet.  In  the  Lebka 
valley  it  made  its  appearance,  as  did  many  other  birds, 
at  a  somewhat  lower  level,  about  2,000  feet,  and  was 
common  at  4,000  to  5,000  feet  in  the  Upper  Lebka 
valley,  and  on  the  Anseba.  Like  all  its  allies,  it  is 
delicious  eating. 

Order  RASOBES. 

Family  PTEEOCLIDiE. 

232.  Pterocles  exustus,  Temm. 

Temm.  PL  Col  364.— Kiipp.  Syst.  Uebers.  No.  384. 

Occasionally  seen  near  the  coast,  but  not  very  common. 
Specimens  killed  were  quite  undistinguishable  from  Indian 
skins. 

233.  Ft.  Lichtensteini,  TemixL 

Temm.  PI.  CoL  355,  361.— Riipp.  Syst.  Uebers.  No.  357. 

Iris  brown ;  orbit  lemon  yellow ;  bill  orange  brown ; 
feet  orange  yellow. 

£  £  2 


420  zooLoor. 

This  bird  has  precisely  tbe  same  habits  as  tl 
dosely  allied  Pt.  fasciatits  of  India.  It  is  rare 
if  ever  seen  on  open  sandy  plains;  like  Pt.  exustu 
it  keeps  to  bush  and  thin  tree  jungle,  and  is  usual] 
found  solitary,  in  pairs,  or  at  the  most  two  or  thn 
pairs  together.  I  once  came  upon  a  considerable  flocl 
in  January,  and  possibly  at  that  time  these  birc 
may  collect  in  large  numbers ;  but  in  May,  June,  Julj 
and  August,  it  was  rare  to  see  more  than  four  togethe 
except  about  watering-places.  When  disturbed  the  Sane 
Grouse  rises  with  a  sharp  cackling  cry,  affording  a  ver 
difficult  shot  It  does  not  rise  high,  and  usually  settle 
again  after  a  short  llight. 

All  kinds  of  Pterocles,  as  is  well  known,  fly  t 
water  at  particular  hours  in  the  day,  the  hours  varjdn 
with  different  species.  Pt.  exustus  drinks  about  9  A.* 
and  4  P.  M.  In  the  presen  t  case  the  drinking-hou 
is  at  daybreak  in  the  morning  and  at  dusk  in  th 
evening,  as  is  also  the  case  with  the  Indian  Pt.  fasci 
atus,  the  crepuscular  Imbits  of  which  are  mention© 
by  Jerdon  ("Birds  of  India,"  vol  iL  p.  498),  and  hav 
been  noticed  by  myself  also.  In  the  semi-desert  countr 
west  and  north-west  of  Massowa,  in  which  Pt.  lAch 
tensteini  abounds,  and  there  are  but  very  few  place 
where  water  is  found,  the  scene  at  each  spring  of  a] 
evening,  after  a  hot  day  especially,  is  very  interesting 
At  Saati,  Ailat,  and  Ain  there  was  a  constant  rush  of  thes 
birds  from  sunset  till  dark,  and  again  in  the  moruinj 
before  sunrise.  Singly  and  in  small  flocks,  uttering  thei 
peculiar  "  queep-queep  "-like  note,  they  flew  up  and  dowi 


AFES,  421 

the  watercourse,  on  their  way  to  or  from  the  water, 
keeping  only  a  few  feet  above  the  bushes  and  low  trees ; 
the  noise  of  their  wings  being  heard  in  the  dusk  before 
the  birds  themselves  appeared.  Like  all  other  Sand 
•  Grouse,  they  are  excellent  eating,  the  flesh  being  rather 
hard,  but  of  delicious  flavour :  and  our  party  used 
generally  to  shoot  a  few  each  evening,  not  an  easy 
matter,  for  the  great  swiftness  and  power  of  wing 
possessed  by  these  birds  rendered  them,  in  the  dusk 
especially,  by  no  means  an  easy  shot. 

Pt.  Lichtensteini  appears  entirely  confined  to  the 
tropical  coast  region.  At  some  water  in  the  Lebka 
valley  at  Mohabar,  only  2,000  feet  above  the  sea,  scarcely 
any  came  to  drink  in  the  evening,  and  at  higher 
elevations  none  were  met  with. 

234.  Pt.  gutturalis,  A.  Smith. 

Smith,  Zool.  South  Afr.  Avos,  pi.  3  and  31. — Riipp.  Syst.  Uehera. 
No.  388. 

Iris  deep  brown. — This  fine  species  was  only  seen  in 
two  or  three  places  on  the  highlands,  and  is  probably 
local,  as  it  is  only  mentioned  in  Ruppell's  lists  from 
Shoa,  though  Von  Heuglin  procured  it  near  Adowa.  I 
shot  a  pair  near  Agula,  drinking  at  about  5  o'clock  in 
the  afternoon,  and  I  also  saw  a  few  close  to  Antalo, 

Family  NUMIDID^ 

235.  Numida  ptilorhyncha,  Licht 

N.  jUilmhijjMJuiy  Licht.  teste  Riipp.  Syst.  Uehers.  p.  102,  t.  xxxix. 

Iris  brown  ;  wattles  bright  verdigris  blue ;  ca.sque  an<l 
the  bristles  above  the  bill  pale  horny ;  legs  reddish,  dusky. 


The  Abyssinian  Guinea-fowl  is  found  thronghoi 
the  country,  from  the  sea-coast  itself  to  an  altitude 
at  least  9,000  feet  These  birds  keep  much  to  cragg 
places,  especially  to  rocky  valleys,  and  often  remain  duiiu 
the  middle  of  the  day  on  the  sides  of  the  steep  or  prec 
pitous  hills.  They  feed  either  in  open  fields  or  in  wooo 
amongst  bushes,  &&,  in  the  morning  and  evening,  an 
roost  at  night  on  high  trees,  a  grove  of  lofty  junipei 
being  frequently  selected  for  that  purpose  in  the  highland 

Throughout  the  winter  and  spring,  the  Guinea-fowJ 
remain  in  large  flocks,  usually  of  200  or  300  birds  eacl 
These  subdivide  into  smaller  flocks  to  seek  food  durin 
the  day,  but  keep  to  one  general  tract  of  country,  an 
unite  again  at  night  Where  not  pursued  they  are  nt 
particularly  wary,  and  but  little  difficulty  is  found  i 
getting  within  gunshot ;  but  along  the  line  of  march  ( 
the  British  army  Uie  Guinea-fowl  soon  became  much  moi 
watchful,  and  in  places  around  Annesley  Bay  where  i 
first  many  were  killed,  it  became  almost  impossible  aft< 
a  time  to  get  near  the  birds,  even  in  cover. 

In  July  and  August  the  flocks  divide  into  pairs,  tw 
or  three  of  which  are  often  found  together,  and  th 
breeding  season  commences.  At  this  time  the  birds  nevt 
appear  to  collect  into  large  flocks ;  I  did  not,  howeve: 
happen  to  see  any  of  tiie  roosting-places.  I  shot 
female  with  a  fully-formed  egg  on  the  9th  of  August.  Th 
young  are  probably  hatched  about  the  end  of  Augvu 
or  beginning  of  September,  aa  they  arc  full-grown  by  th 
end  of  the  year. 

The  voiec  is  very  similar  to  that  of  the  common  tam 


AVES.  423 

Guinea-fowl.  The  food  appears  to  consist  to  a  larger  extent 
of  seeds  and  fruits  than  amongst  the  Partridges,  insects 
being  apparently  but  little  sought  after.  In  one  instance 
three  birds  shot  one  morning  near  Halai  had  been  feeding 
chiefly  upon  the  small  tubers  or  conns  of  the  Quentee 
{Cyperus  esculentus).  Their  crops  also  contained  seeds 
and  a  few  fragments  of  leaves,  but  amongst  the  three 
only  one  insect,  a  Ilemipter.  Their  being  so  exclusively 
gramnivorous  is  probably  the  cause  of  the  excellence  of 
their  flesh,  which  is  far  superior  to  that  of  most  game 
birds.  As  may  be  imagined,  they  were  much  sought 
after  by  the  sportsmen  of  the  army,  and  during  the  stay 
of  our  small  party  in  the  Anseba  valley  we  lived  to  no 
small  extent  upon  these  birds,  as  their  flesh  was  as  far 
superior  to  that  of  all  other  game  except  the  Koodoo,  ba 
it  was  to  that  of  the  tough  goats  which  we  could  obtain 
not  without  difficulty  from  the  villagers. 

Family  PERDICIDiE. 

236.  Francolinus  Erkelii  (Hiipp.). 

Perdi<c  Erkdii,  Riipp.  Neu.  Wirb.  p.  12,  t  vi. 
Frankolimu  Erkdii,  Riipp.  Syst.  Uebers.  No.  378. 

Iris  hair  brown ;  beak  black ;  legs  yellow  with  much 
dusky  on  the  toes ;  claws  dusky. 

This  is  perhaps  the  commonest  Partridge  of  the 
Abyssinian  highlands.  It  keeps  much  to  rocky  hill- 
sides, and  has  a  considerable  range  in  altitude,  from 
about  3,000  feet  above  the  sea  to  10,000  feet.  It 
abounded  in  the  upper  portion  of  the  Scnafe  pass,  and 
from  the  male  being  provided  with  two  strong  spurs. 


424  ZOOLOOY. 

it  was  very  generally  known  in  the  army  as  SpuHowl, 
the  name  by  which  GcUloperdix  spadicetis,  the  male 
of  which  is  also  double-spurred,  is  generally  known  in 
India.  The  two  forms  are  not  by  any  means  closely  allied, 
but  the  name  is  far  less  objectionable  than  that  of  Pheasant 
applied  by  the  Cape  colonists  to  a  species  closely  allied 
to  F.  JSrkelii.  In  the  Anseba  at  4,500  feet  F.  ErkeUi 
was  only  occasionally  seen,  but  it  inhabited  the  hills 
above  the  valley. 

This  Francolin  is  highly  insectivorous,  and  the  Hesh, 
although  generally  excellent,  has  occasionally  a  rather 
strong  taste,  due  doubtless  to  its  food.  Some  which  I 
killed  about  Undul  in  January  had  the  crops  fOled  with 
locusts.  The  grain  dropped  from  the  bags  in  the  pass 
was  a  great  9*ttraction,  and  many  might  be  seen  late  in 
the  evening  or  eajrly  in  tfee  morning  picking  it  from  the 
road  through  the  pass.  Just  below  Senafg,  I  have  seen 
fifteen  or  twenty  of  these  birds  in  a  walk  of  a  mile.  On 
being  disturbed  they  usually  ran  up  the  hill,  very 
rarely  down,  and  never  took  flight,  unless  fired  at  or 
suddenly  surprised.  Aboi;it  dusk  in  tiie  evening  they 
roosted  in  trees,  and  the  whole  hill-side  resounded  with 
their  cries  "  kri-kri-kri-kri-kri-ka-wft-warwftrwft,"  a  harsh 
sound,  the  first  syllables  lengthened,  the  others  quickly 
uttered. 

I  never  saw  Francolinus  Erkelii  in  coveys ;  it  was 
usually  in  pairs,  or  single.  The  difierence  in  size  between 
the  sexes  is  very  great,  far  more  than  is  common  amongst 
Partridges.  The  following  are  measurements  in  inches  of 
a  pair,  taken  in  the  flesh  :— 


AFES.  425 


Male  . 
Female 


Length. 

Wing. 

OML 

Tanu.  Beakinfhmt. 

in. 

in. 

in. 

in.              in. 

19 

8 

5 

n      u 

16 

9 

4i 

2*            1* 

237.  Fr.  Riippelli,  G.  R  Gray. 

Gen.  Birdfl,  iiL  p.  605. 

Ferdix  Clappertoni,  l^iipp,  (|iec  Childr.)  Atlas,  t,  «. 

Francolinus  Biippclli,  Biipp.  Sjst.  Uebers.  No.  379. 

Iris  brown  ;  naked  skin  around  the  eye  and  legs  deep 
red ;  the  male  with  a  double  spur ;  bijl  black. 

This  bird,  though  found  in  the  temperate  region,  and 
common  in  places  about  Antalo,  rather  belongs  on  the 
whole  to  a  lower  zone  than  F.  JErkelii  and  F.  gutturcUis. 
I  never  met  with  it  in  the  Senafd  paas,  nor  in  the  province 
of  Agamd  about  Sejiafd  or  Adigrat,  nor  in  the  higher 
valleys  further  south,  whilst  it  was  most  abundant  in  the 
Upper  Lebka  and  Anseba  valleys  fA  from  4,P00  tp  5,000 
feet.     The  cry  is  harsh  and  monotonous, 

238.  Fr.  gutturalis,  Eiipp. 

Riippell,  Sjst  Uebers.  p.  103,  t.  xL 

Iris  brown ;  legs  yeUowish  brown, 

In  Decepiber,  January,  and  February  these  birds  were 
met  with  in  small  coveys ;  subsequently  they  were  seen 
in  pairs,  generally  amongst  bushes  in  yalleys^  and  not 
keeping  to  rocky  hill-sides  like  the  last  species.  They 
were  not  seen  in  the  pass,  but  were  common  around 
Senaf(6,  and  moderately  so  throughout  the  highlands.  In 
the  Anseba  valley  they  were  less  abundant  than  the  next 
species,  but  were  more  common  in  the  valley  itself  than 
F.  ErkeliL     At  this  time  in  July  and  August  the  flesh 


426  ZOOLOGY. 

was  sometimes  bo  rank  as  to  be  scarcely  eatable,  doubtiess 
from  their  having  fed  largely  on  the  Golec^tera  which 
then  abounded.  In  the  winter  months  they  were 
excellent. 

The  call  is  very  similar  to  that  of  the  common  English 
Partridge,  to  which  the  plumage  also  presents  some 
resemblance,  so  that  sportsmen  often  take  them  for  the 
same  bird.  In  reality  the  present  species  rather  resembles 
the  common  grey  Partridge  of  India,  Francolinus  {Orty- 
giornis)  pondicerianus. 


339.  Ptemistes  rubricollis  (Kiipp.). 

Perdix  ruiricollit,  Btipp.  Atba,  t  ixx. 

FUmUUt  TubrieoUu,  Eupp.  Syst.  Ueben.  Ho.  38S. 

Iris  brown ;  legs  reddish,  dusky ;  beak  dusky ;  gape 
reddish ;  naked  skin  around  the  eyes,  with  the  lores,  part 
of  the  ear,  chin,  and  throat  scarlet,  passing  downwards 
into  yellow  on  the  neck. 

This  is  the  Partridge  of  the  coast,  and  was  never  seen 
above  3,000  feet  It  was  not  observed  above  Eomayli 
at  the  bottom  of  the  pass  leading  to  Senafd.  In  Decem- 
ber and  January  it  occurred  in  coveys  of  sis  to  eight 
birds,  subsequently  single  or  in  pairs.  It  chiefly  keeps 
about  bushy  ravines,  and  is  a  great  runner,  not  taking 
wing  readily,  but  generally  when  disturbed  endeavouring 
to  run  away.  It  has  a  peculiar  harsh  cry,  uttered  when 
disturl>ed ;  I  did  not  notice  its  call.  Like  Francolinus 
Erkelii,  it  was  generally  known  by  the  sportsmen  of  the 
army  as  Spur-fowl. 


AFES.  427 

Oedeb  GR  all  a  tores. 
Family  OTIDiK 

240.  Otis  (Eupodotis)  arabs,  L. 

Linnaeus,  Syst.  Nat.  ed.  12*,  i.  p.  264. — RiippeU,  Atlas,  t.  xvi; 
Syst.  Uebers.  No.  319. 

Tris  pale  brown,  with  rather  irregular  dark  streaks 
radiating  from  the  pupil ;  bill  dusky  above,  dirty  white 
below ;  legs  pale  dirty  yellow. 

This  fine  Bustard  is  common  on  the  plains  near  the 
coast,  in  open  places,  or  in  thin  jungle,  keeping  a  good 
deal  to  the  former.  The  tracks  were  seen  at  an  elevation 
of  about  3,000  feet  at  Rairo  in  Habab,  but  a3  a  rule  it 
appears  confined  to  the  plains.  Like  most  Bustards,  it  is 
rather  wary,  and  is  much  easier  to  approach  on  horseback 
than  on  foot.  I  shot  several  from  the  saddle.  The  flesh 
is  very  good. 

In  the  stomach  I  found  locusts  and  a  few  Cohoptera, 
The  male  is  much  larger  than  the  female,  as  usual  in  this 
sub-genus  of  Bustards. 

241.  O.  (Sypheotides)  melanogaster,  Eiipp. 

Otu  melanogaster,  Eiipp.  Neu.  Wirb.  p.  16,  t.  vii.  (mas.) ;  Syst. 
Uebers.  p.  107,  t.  xli. 

Though  this  small  Bustard  was  common  on  the  table- 
land in  places,  especially  on  the  open  plains  between 
Adigrat  and  Antalo,  I  never  shot  a  specimen  myself,  and 
for  the  two  obtained  I  am  indebted  to  Lieut.  St.  John. 
Many  were  killed  by  oflicers  along  the  line  of  march. 
They  occurred  in  grass  and  bushes,  generally  lying  close, 


I 


:; 


1 


I  1 


i 


428  ZOOLOGY. 

and  hiding  as  the  Floriken  of  India  does.  Indeed  this 
bird  is  a  true  Floriken  in  all  its  habits,  and  the  plumage 
of  the  male  undergoes  a  similar  change  to  that  of  the  two 
Indian  species  of  Sypheotides,  being  grey  in  the  winter 
and  becoming  in  great  measure  black  in  the  spring. 
Specimens  shot  in  May  had  assumed  the  black  plumage 
to  a  great  extent  The  flesh  is  delicious,  quite  equal  to 
that  of  the  Indian  Floriken,  which  is  one  of  the  most 
delicately-flavoured  game  birds,  apd  far  superior  to 
the  large  Bustarda 

Family  CHARADRIDiE, 


'  i  .  242,  /^dicnemus  crepitans,  Ternm. 

if 

Riipp.  Syst  Uebeis.  No.  395. 


(  Common  near  the  coast.     I  cannot  see  any  difierence 

from  European  specimens. 


i< ;  243.  JE.  afiinis,  Eiipp. 

Biipp.  Mu8.  Senck.  ii,  p.  310;  Srst.  Uebers.  p.  lU,  t.  xlji. 

I  only  saw  this  bird  twice  or  three  times,  and  I  killed 
lijl?  in  August  a  single  specimen  in  the  Anseba  valley,  into 

which  it  appeared  to  be  migrating,  as  I  saw  none  before, 
8  J  but  both  I  and  Mr.  Jesse  saw  it  there  more  than  once 

just  before  leaving.    It  appeared  to  have  precisely  the 
same  habits  as  the  common  Stone  Curlew. 

The  figure  of  this  bird  in  Ruppell's   "  Systematische 
Uebersicht "  is  unsatisfactory.     This  is  the  only  instance 
in  the  book,  I  believe,  as  most  of  the  plates  are  admir- 
able, and  give  an  excellent  representation  of  the  birds 
I  fisfured. 


^ 


AFES.  429 

244.  Eudromias  asiaticus  (Pall.). 

Apparently  less  abundant  than  the  last  on  the  coast. 
I  shot  some  specimens  inland  at  Rairo  in  Habab  in 
August     They  were  in  flocks,  on  open  grassy  ground. 

245.  ^gialitis  GeofFroyi  (Wagler). 

Chouradrius  Geoffroyi,  WagL  Syst.  Ayinm,  Charadrios  19,  1827. 
0.  LeschenatiUiiy  LeBson,  Manual  d'Om.  i.  p.  22%  1826. 
Hiaticula  Oeoffroyi,  Riipp.  Syst  U«bers.  No.  415. 

Common  on  the  sea-coast.  I  killed  one  specimen  with 
a  red  breast  in  January. 

246.  JE.  niveifrons,  Cuv. 

This  was  abundant  at  ZuUa  in  June,  and  apparently 
breeding,  for  the  birds  ran  about  with  the  usual  tricks, 
pretending  lameness  and  inability  to  fly.  I  also  shot  a 
specimen  at  Massowa  in  August. 

247.  JE.  tricoUaris,  YieilL 

Cha/radr%u$  iricollariif  Yieillot,  Noiiv^  Diet   d^Hist.  Nat  xxvil 

p.  147. 
Hiaticula  indica,  Rtipp.  Syst  Ueben.  No.  41 2^ 

Iris  grey,  darker  outside,  orbit  orange  red ;  bill  with 
basal  half  flesh-coloured,  tip  black ;  legs  livid  flesh- 
colour. — In  flocks  about  freshwater  springs  at  Ailat  and 
Amba  near  Massowa,  never  seen  on  the  sea-coast. 

248.  Chettusia  melanoptera,  Eiipp. 

Iris  yellow,  greenish  round  the  pupils ;  skin  round 
the  eye  and  legs  purplish  red. 

In  the  characters  of  the  wing  and  in  the  absence 
of  the  hind  toe,  this  form  belongs  to  the  Charadritia 


J 

! 

ll  430  ZOOLOGY. 

I'  group,  while  its  voice,  habits,  and  flight  are  unmistake- 

[>  ably  those  of  a  Lapwing.     In  voice  and  flight  it  resembles 

Hoplopterus.  It  is  found  usually  in  small  flocks  on 
grassy  meadows  throughout  the  highlands,  ascending 
above  10,000  feet.  I  never  met  with  it  below  about 
7,000  feet ;  indeed  the  small  meadows  along  the  banks 
of  streams  in  which  it  is  generally  seen  are  not  common 
below  the  temperate  region  in  Abyssinia. 

249.  Sarciophorus  tectus  (Bodd). 

Charadrius  pileatuSf  Gm.  Syst.  Nat  L  p.  691. 
Sarciophorus  pUecUus,  Riipp.  Syst  Uebers.  No.  408. 


ii  * 


f  ]  Iris  yellow,  with  a  dusky  circle  round  the  pupil ;  bill 

i ,  and  small  wattle  dull  crimson  ;  tip  of  bill  dusky ;  legs 

f  J  purplish  red. 

;  f  I  This  is  a  lowland  form,  chiefly  confined  to  the  coast 

country,  but  I  met  with  it  in  the  Anseba  valley. 


An 


/ 


f, 


' 


I 


250.  Lobivanellus  melanocephalus,  Eiipp. 
Biippell,  Syst  Uebers.  p.  116,  t  xliv. 

Iris  grey,  darker  towards  the  pupil ;  skin  round  the 
eyes  orange,  paler  externally;  bill  black;  wattle  and 
fleshy  portion  of  base  of  upper  mandible  pale  lemon 
yellow ;  legs  yellow. 

This  species  of  Plover  was  only  seen  on  the  higher 
portion  of  the  Dalanta  and  Wadela  plateaux,  above  10,000 
feet  elevation.  Rlippell  met  with  it  on  the  mountains 
of  Simen  in  April,  when  I  saw  it.  It  was  in  large 
flocks,  which  were  very  wary.  It  is  a  noisy  bird,  with 
a  shrill  call  like  "  che-wip.''  The  following  measurements 
were   taken  in  the  flesh : — Length  13^  in.,  extent  30, 


AVES.  431 

closed  wing  9|,  tail  4,  tarsus  2|,  middle  toe  Ig,  beak 
from  front  ItV. 

251.  L.  senegallus  (L.). 

Parra  senegcUla,  L.  Syst.  Nat.  ed.  12»,  I  p.  259. 
Lobivanelltts  serugahrmSy  Biipp.  Sjst  Uebers.  No.  406. 

Iris  grey,  clouded  towards  the  pupil ;  skin  round  eye 
yellow ;  wattle  triangular,  pointed  below,  stretching  from 
the  eye  to  the  beak,  the  upper  portion  red,  lower  yellow, 
the  two  colours  distinct,  not  blending ;  beak  yellow,  except 
the  tip  of  the  upper  mandible,  which  is  black ;  legs 
greenish  yeUow. 

This  Lapwing  was  only  seen  on  the  highlands  near 
Agula  below  7,000  feet,  and  in  the  Anseba  valley,  and 
was  evidently  scarce  in  both  localities. 

252.  Hoplopterus  spinosus  (L.). 

Biipp.  Syst.  Uebers.  No.  407. 

Iris  crimson  ;  legs  and  bill  black. 

Only  seen  on  the  highlands  on  the  banks  of  Lake 
Ashangi,  but  it  also  occurred  at  Ailat,  and  on  the  edges  of 
a  small  stream  at  Amba  near  Massowa.  Like  the  Indian 
species  H.  ventralisy  it  is  probably  mainly  foimd  on  river 
banks,  and  never  away  from  fresh  water.  The  cry  is 
precisely  similar  in  both  races. 

253.  Glareola  pratincola,  L. 

G,  limbata,  Biipp.  Syst.  Uebers.  p.  113,  t.  xliiL 

I  obtained  two  specimens  at  Rairo  in  Habab,  at  about 
1,000  feet  above  the  sea,  on  a  grassy  plain  far  from  any 
water.     There  is  no  question  but  that  Etippell's  species 


was  founded  on  a  young  bird.  The  Eastern  form  6. 
orientalis,  distinguiEihed  by  having  the  tail  square  and 
not  forked,  does  not  appear  to  be  found  so  far  west  aa 
the  coasts  of  the  Red  Sea 

S54.  Dromas  ardeola,  Payknll 

Pft;kiiU,  Haod.  k.  Vatenk.  Acad.  Stockholm,  ISOS,  ^  108,  t.  viii. 

Id.  Btipp.  Syst.  Uebere.  No.  430.' 

Brodia  ampKiUntis,  Salt,  App.  p.  IxL  ud  plate. 

Not  rare  on  the  coast,  usually  in  small  flocks. 


Faidlt  8C0L0PACID.S. 
955.  Numenius  arquatus.  L. 

Riipp.  STst.  Uehon.  Na  402. 

Common  on  the  shore  in  winter. 

356.  N.  ph«opu8,  L. 

Eiipp.  Syst  tXebeii.  No.  463. 

A  single  specimen  obtained  at  Massowa  on  August  25. 

957.  Scolopaz  gallinago,  L. 

Run>-  SM  Uebeta.  Na  470. 

Common  on  the  highlands  in  the  winter  months. 
There  were  many  in  marshy  spots  on  the  shores  of  Lake 
Ashangi ;  as  late  as  the  beginning  of  May,  eight  to  ten 
couple  were  shot  by  a  single  officer  on  two  afternoons, 
in  about  a  couple  of  hours'  time. 

358.  Rhynchsea  bengalensis  (L.> 

Biipp.  STst.  Uebers.  No.  469, 

The  Painted  Snipe  was  only  met  with  in  reeds  on 
the  banks   of  running   water   at  Ailat    and  Ain,  near 


AFES.  433 

Massowa.  Only  two  or  three  birds  were  seen  at  each 
place.  The  specimens  are  quite  undistinguishable  from 
Indian  ones. 

259.  Terekia  cinerea  (Gm.). 

A  single  specimen  shot  at  Zulla  in  January. 

260.  Actitis  hypoleucos  (L). 

Totanus  hypoleucos^  Biipp.  Syst.  Uebers.  No.  468. 

Common  near  inland  streams.    I  only  collected  speci- 
mens on  the  Anseba. 

a6i.  A.  ochropus  (L.). 

Riipp.  Syst.  Uebers.  No.  460. 

Common  on  the  highlands.     I  did  not  note  its  occur- 
rence on  the  coast. 

262.  Totanus  calidris,  Bechst. 

Scolopax  calidris,  Salt,  No.  52,  App.  p.  xlviii. 
jT.  calidrisy  Riipp.  Syst.  Uebers.  No.  459. 

A  single  specimen  obtained  on  the  coast  in  January. 

263.  Tringa  cinclus,  L. 

Obtained  at  the  same  place  and  about  the  same  time 
as  the  last. 

264.  T.  minuta,  Bechstein. 

Also  shot  at  Zulla. 

Family  RALLIDiE. 

265.  Rallus  Rougeti,  Gu^r. 

Gu4riu,  Rev.  ZooL  1844,  p.  322.— Ferr.  et  Gal.  iii.  p.  253. 
B.  abyannicuSy  Riipp.  Syst*  Uebers.  p.  127,  t.  xlvi. 

Iris  dull  red ;  beak  also  dull  red ;  legs  reddish  brown. 
Common  amongst  reeds  and  bushes  in  swamps,  and  on 

F  F 


i 


434  aOOLOGY. 

the  banks  of  streams  in  the  highlands.  It  has  the  usual 
habits  of  the  genus,  being  a  Bkulkiug  shy  bird,  not  eai^  to 
shoot,  from  its  always  hiding  in  the  thickest  bush  or  tuft 
of  grass,  and  from  its  unwillingness  to  take  flight.  It 
has  a  very  loud  harsh  croaking  call,  repeated  severa] 
times  in  succession.  On  the  borders  of  Lake  Ashangi, 
in  some  swamps  with  rushes  about  three  feet  high,  there 
were  many  of  these  Rails.  In  the  day-time  they  were 
invisible ;  one  might  walk  about  the  marshes  for  houn 
without  suspecting  their  presence ;  but  about  sunset  thej 
used  to  sit  on  the  tops  of  the  rushes  and  caU.  At  a  shot 
all  vanished  amongst  the  reeds,  but  they  soon  reappeared 
By  noticing  this  habit  I  obtained  several  Erpecimens. 
a66.  Galltnula  chloropus,  L. 
Kiipp.  Sf  bL  Ueben.  No.  479. 

The  common  Moor-hen  was  pretty  common  around 
Lake  Ashangi,  and  I  occasionally  saw  it  elsewhere. 

367.  Fulica  cristata,  L. 

Iris  didl  red ;  tegs  olive-grey  ;  beak  and  forehead  flesh- 
coloured  ;  the  two  fleshy  knobs  on  the  occiput  dull  r«d, 
and  much  larger  in  the  male  than  in  the  female. 

The  Crested  Coot  precisely  resembles  the  Common  Coot 
in  its  habits,  and  abounds  on  Lake  Ashangi 

Faidlt  AEDEID^E. 
a68.  Ardea  atricollis,  WagL 

WagL  Syat.  At.  Ardea,  ap.  4 

.  8a— Hengl  SyeL  Uebere. 


Iris  bronze  yellow  ;  upper  mandible  dusky,  lower  pale  ; 
legs  dusky ;  soles  of  feet  yellowish. 


AFES.  436 

I  only  once  obtained  a  specimen  of  this  Heron.  It  was 
killed  in  a  swampy  meadow  near  Adabagi,  two  marches 
south  of  Adigrat,  at  an  altitude  of  about  8,000  fipet  above 
the  sea. 

269.  Herodius  garzetta,  L.  yar, 

Riipp.  Syst  Uebers.  No.  427. 

Common  on  the  coast  I  also  saw  an  Egret  on  the 
highlands,  probably  the  same,  but  I  failed  to  secure  a 
specimen. 

270.  Ardea  (Demi-egrttta)  gularis,  Bosk. 

Act  Soa  d*Hist.  Nat.  L  p.  4,  pi.  il 
A,  8chistaceay  Hemp,  and  Ehr.,  Symb.  PhyfL  t.  vi 
A.  asha,  Sykes,  P.  Z.  S.  1832,  p.  157. 
Egretta  gularis,  Eiipp.  Syst  Uebers.  No.  428*. 

Iris  yellowish  white ;  bill  reddish  yellow ;  legs  black ; 
feet  yellow. 

I  obtained  a  single  specimen  at  Dissee  Island.  As 
shown  by  Mr.  Blyth  (Joum.  As.  Soc.  Bengal,  1855),  there 
can  be  no  question  of  the  identity  of  this  species  with 
the  Indian  form.  I  have  compared  my  specimen  care- 
fully with  those  in  the  Calcutta  Museum. 

271.  Ardea  {Ardeola)  comata,  Pal). 

Pallas,  Reise,  ii.  p.  715. 

A.  ralloides,  Riipp.  Syst  Uebers.  Na  432. 

I  obtained  a  specimen  at  Antalo.  This  bird  was  only 
occasionally  seen  on  the  banks  of  freshwater  streams. 

272.  Ardea  {Butorides)  atricapilla,  AfzeL 

Afzel.  Of  Stockh.  1804. 

I  shot  a  specimen  in  winter  plumage  in  February,  and 
another  in  the  breeding  livery  in  August,  both  on  the 
searshore  amongst  mangrove. 

F  F  2 


FiKiLT  CICONID^ 

373.  Scopus  umbretta,  Gm. 

Kiipp.  Syrt.  tJeboM,  No.  430. 

This  peculiar  bird  was  not  common  in  general  on  th( 
higUandB,  although  a  few  were  seen  about  I^ake  Asbang; 
and  elsewhere.  It  was  abundant  on  the  Anseba,  afibei 
the  rains  had  fairly  Bet  in. 

374.  Ciconia  abdimii  {Hemp,  and  Ehi.). 

Sphmorhynthiu  aidimii,  Hempt  and  Ehr.,  Symb.  Vhjt.  t.  t. 
Cietmia  abdimii,  Riipp.  AQm,  L  viii. ;  Syrt.  Ueben.  No.  443. 

Irla  dusky  grey ;  beak  duaky  green,  reddish  at  the  gapt 
towards  the  tip ;  naked  skin  of  the  sides  of  the  head 
lavender,  deep  red  in  front  of  the  eye  and  under  th( 
chin,  white  round  the  back  of  the  orbit ;  ears  red ;  the 
akin  beneath  the  feathers  red  throughout ;  feet  duU  red ; 
legs  olivaceous. 

Occaaionally  seen  on  the  highlands,  but  not  common. 
About  the  shores  of  Annesley  Bay  it  is  much  more 
frequent,  and  I  several  times  saw  it  in  the  Anseba  valley. 
It  is  by  no  means  a  shy  bird  in  general,  but  sometimea 
difficult  to  approach  when  on  open  plains.  I  once  saw 
five  by  some  carcases  of  mules  at  Komayli,  doubtleSa 
attracted  by  the  insects. 

Family  T&KIALXDM 

275.  Ibis  {Geronliciu)  comata,  Ehr. 

Gerontiew  amatui,  Ehr.  apnd  Riipp.  Sj«t  Uebera.  p.  IftS,  t  xlv. 

I  only  met  with  this  bird  on  the  highlands,  once  near 
Senafe,  and  again  in  a  large  flock  near  Antalo.    On  the 


AVl^.  437 

second  occasion  I  secured  two  specimens.     It  appears  to 
be  by  no  means  common  in  the  region  traversed. 

276.  I.  (Harpiprion)  carunculata,  Eiipp. 

Biipp.  Neu.  Wirb.  p.  49,  t  xiz. ;  Syst.  Ueben.  No.  448. 

Iris  pearly  white,  with  a  broad  brown  ring  round  the 
pupil. 

This  is  a  far  more  common  bird  than  the  last, 
abounding  throughout  the  higher  portions  of  the  high- 
lands. It  was  common  around  Senafe,  at  Ashangi  in  the 
meadows  and  swamps  around  the  lake,  and  above  all  on 
the  Wadela  and  Dalanta  plateaux.  It  is  seen  solitary  or 
in  small  flocks,  usually  keeping  to  open  meadows  or 
cultivated  fields,  and  feeding  upon  snails,  caterpillars, 
beetles,  locusts,  &c.     It  has  a  harsh  cry,  seldom  uttered. 

Order  NATATOBES. 

Family  ANATEDigE. 

277.  Anas  flavirostris,  Smith. 

BiU  yellow ;  nail  at  end,  a  stripe  between  the  nostrils, 
and  all  the  basal  part  of  the  lower  mandible  black  ;  legs 
blackish  brown. 

Common  throughout  the  highlands,  and  especially  on 
the  higher  plateaux,  in  all  streams. 

Measurements  in  the  flesh :  length  21  in.,  closed 
wing  10^,  tail  3^,  bill  from  front  If  in. 

278.  Fuligula  cristata,  Bay. 

Lake  Ashangi,  in  pairs,  and  not  very  common  at  the 
commencement  of  May. 


438  ZOOLOGY. 

979.  Querquedula  crecca,  L 

Occasionally  met  with  on  the  highlands.  I  did  not 
eee  Q.  circia. 

aSo.  Chenalopex  Bgyptiaca,  L. 

Kiipp.  STBt.  Uebers.  No.  487. 

Iris  orange  yellow ;  bill  flesh  colour  above  and  below  ; 
tips  and  edges  of  both  mandibles  and  base  of  the  upper 
one  dusky,  but  the  dark  colour  is  unequally  distributed 
in  difieient  E^cimens,  and  is  sometimes  wanting ;  legs 
pink,  with  pale  mottling. 

A  common  biid  throughout  the  highlands.  It  is  in 
general  excellent  eating,  and  was  consequently  much 
sought  after  by  the  sportsmen  of  the  army ;  and  a  large 
plain,  about  four  miles  south  of  Senaf^,  on  which  the 
3d  Bombay  Cavalry  were  for  some  time  encamped,  was 
known  as  "  Gooae  Plain,"  from  the  numbers  of  these 
birds  found  upon  it.  It  keeps  chiefly  to  open  grassy 
plains  near  streams,  during  the  winter  and  spring 
months  at  least,  and,  like  most  Shieldiakea,  appears  to 
keep  much  in  pairs  at  all  seasons  of  the  year,  these  pairs 
occasionally  collecting  into  considerable  societies,  which 
however  have  by  no  means  the  same  united  volition 
which  may  be  seen  in  birds  habitually  fotmd  in  flocks, 
lUce  most  Ducks  and  Geese.  On  the  shores  of  Lake 
Ashangi  the  Egyptian  Goose  abounded,  and  hundreds 
might  be  seen  scattered  over  the  grassy  plains  around 
the  lake ;  but  they  came  to  the  feeding-grounds  singly 
or  in  pairs,  rarely  in  larger  numbers,  and  flew  away  in 
tlie  same  manner. 


ArES.  439 

About  the  middle  of  July  several  of  these  birds  made 
their  appearance  in  pairs  along  the  banks  of  the  Anseba^ 
then  a  muddy  torrent,  and  commenced  building  on 
the  trees  overhanging  the  stream.  Mr.  Jesse  killed 
one  hen  bird  at  the  end  of  July,  containing  a  fully- 
developed  egg. 

In  most  of  its  habits  this  bird  closely  resembles 
the  common  Chakwa  of  India,  Ccbsarca  rutilanSy  but 
it  is  more  frequently  seen  on  meadows.  Both  its  flight 
and  call  are  those  of  a  Shieldrake  rather  than  of  a 
Goose. 

The  brown  chest-spot  varies  much  in  dimensions,  and 
is  sometimes  wanting.  It  is  usually  larger  in  the  males 
than  in  the  females. 

281.  Bemicla  cyanoptera,  Eiipp. 

Riippell,  Syst.  Uebeis.  p.  129,  t  xlvil 

This  dwarf  Goose  abounds  on  the  Wadela  and  Dalanta 
plateaux,  above  9,000  to  10,000  feet,  and  is  a  conspicuous 
member  of  the  Subalpine  fauna  to  which  reference  has 
so  frequently  been  made.  I  only  saw  the  bird  once 
elsewhere  ;  this  wa^  ii^  a  high  valley  at  about  8,500 
feet  above  the  sea,  between  Antalo  and  Ashangi  The 
habits  of  the  bird  are  rather  those  of  a  Shieldrake  than 
of  a  Goose,  for  it  is  not  met  with  in  flocks  (at  least  that 
is  not  the  case  in  April,  the  season  when  I  saw  it),  but 
singly  or  in  pairs,  walking  about  on  open  grass  plains 
near  streams  and  swampa 


440  Z00L03T. 

-SaoLt  PODICIPID^ 
383.  Podiceps  cristatus,  L. 

Heoi^  Sfst  Uebera.  No.  706. 

Iris  Indian  red ;  beak  dusky  reddish  brown ;  legs  and 
feet  dusky  olive ;  dirty  yellow  on  the  webs,  and  on  the 
upper  part  of  the  tarsus. 

The  Crested  Grebe  is  extremely  abundant  on  Lake 
Ashangi,  where  it  is  doubtlese  a  permanent  resident. 

383.  P.  auritus,  L. 

HeogL  Syrt.  Uebws.  No.  708. 
Not  very  common  on  Lake  Ashangi.  It  is  curious 
that  Uie  people  in  the  neighbourhood  persist  in  declaring 
that  there  are  no  fish  in  this  lake,  yet  it  abounds  in  fish- 
eating  birds.  I  saw  shoals  of  small  fish,  but  had  no 
means  of  catching  them ;  and  I  have  httle  doubt  that 
larger  kinds  also  occur,  for  I  found  tracks  of  Otters  along 
the  shore,  and  several  were  seen  by  a  Mend. 

384.  P.  minor,  lath. 

Biipp.  SjBt.  Uebera.  No.  502, 

Not  abundant  on  Lake  Ashangi  I  also  saw  it  once  on 
a  pool ;  and  one  species,  perhaps  the  same,  occurred  on 
the  coast  at  Zulla. 

Familt  LAKID^ 

385.  Larus  fuscus,  L. 

Common  at  Zulla.  The  wing  is  a  httle  longer,  about 
an  inch,  and  the  bill  a  trifle  smaller,  than  in  most 
European  specimens,  but  there  appears  to  be  no  constant 
distinction. 


AFJSS.  441 

286.  L.  Hemprichii)  Bonap. 

Perhaps  the  commonest  species  in  Annesley  Bay.  It 
also  abounds  in  Aden  harbour. 

287.  L.  leucophthalmus,  Licht. 

Temm.  PL  CoL  366. 

# 

Xema  lettcophthcUmum,  Biipp.  Syst  Uebers.  No.  507. 

Common  in  Annesley  Bay. 

288.  Sterna  (Thalasmu)  affinis,  Eiipp. 

Riippell,  Atlas,  t.  xiv.  1826 ;  Syst  Uebers.  No.  618. 
8,  bmgcUensiSy  Lesson,  Traits  d'Omith.  p.  621  (1831). 

Abundant  on  the  sea-coast  near  Zulla.  Frequently 
seen  in  large  flocks. 

289.  Sterna  {Onychoprion)  panayensis,  Gm. 
Iris  dark  brown ;  biU  and  legs  black. 

I  did  not  meet  with  this  species  in  Annesley  Bay,  but 
it  abounded  farther  south  in  the  Red  Sea,  and  I  captured 
several  specimens,  which  alighted  on  the  boat  during  my 
voyage  from  Massowa  to  Aden. 

Family  PHAETONlDiE. 

290.  Phaeton  rubricauda,  Bodd. 

A  single  specimen  of  a  young  bird  was  captured  alive 
in  Annesley  Bay,  and  sent  to  me  by  Captain  Mockler. 
I  did  not  notice  any  flying  about  the  bay.  The  Tropic 
Birds  usually  keep  out  at  sea. 

Family  GRACULIDiE. 

291.  Graculus  africanus,  Gm. 

Iris  crimson ;  bill  flesh  colour,  dusky  on  culmen ;  feet 
black. 


I  ouly  saw  Cormorants  on  Lake  Ashangi,  and  onl 
this  species.     It  is  not  very  ftbundant. 

Family  PELICAHID^ 

293.  Pelicanus  nifescens,  Latti. 
Riipp.  Atlu,  t.  ii. 

Iris  yellowish  brown,  mottled  darker ;  bill  yellowis 
white ;  nail  at  tip  of  upper  mandible  orange ;  pouch  fles 
colour,  with  fine  transverse  equidistant  and  parallt 
yellow  lines;  legs  yellowish  white. 

I  shot  a  pair  of  these  birds  on  Dissi  Island,  at  the  en 
of  Augiist.  They  were  in  fine  plumage,  with  a  ric 
roseate  tinge  on  the  back, 

393.  P>  phiUipensis,  Om. 

Iris  grey  brown  ;  bill  flesh  colour ;  nail  at  end  of  uppe 
mandible  yellow ;  gular  pouch  livid ;  legs  and  feet  pal 
flesh  colour. 

Between  this  bird — which  was  common  on  the  shor 
at  ZuUa,  and  of  which  I  shot  a  single  specimen — an' 
the  last  there  is  no  difierence  in  measurements,  h 
the  form  of  the  frontal  feathers,  or  the  disposition  0 
those  behind  the  eye.  The  crest  also  appears  similar 
and  although  the  feathers  of  the  head  are  shorter  am 
more  woolly,  this  may  partly  be  due  to  wear.  Unde 
these  circumstances,  T  should  have  been  inclined  to  con 
sider  the  roseate  back  and  breast  of  P.  rufescens  a 
seasonal,  but  for  the  circumstance  that  while  the  presen 
specimen  is  unquestionably  identical  with  the  commo] 


111  ^ 


AVES.  443 

Indian  P.  phillipensis,  with  specimens  of  which  I  have 
compared  it,  P.  ru/escens  does  not  appear  to  have 
been  noticed  in  India.  If  it  were  only  the  nuptial 
plumage  of  the  present  species,  it  ought  to  be  equally 
common. 


444 


ZOOLOGY. 


H 


li  / 


/ 


I , 


Class    REPTILIA.' 
Obdbr  CHELONIA. 

1.  Cinixys  belliana,  Gray. 

Kinixyi  beUiana,  Gray,  Synopsis  Sept.  Add.  and  Coir.  p.  i 
Catalogue  of  Shield  Bept  in  ColL  Brit  Mus.  1855,  p.  13, 
Cinixys  heUiana,  Dum.  et  Bibr.  Eip.  Gen.  ii  p.  168. 
Kinixys  SkoMxms^  Biipp.  Mus.  Senck.  ilL  p.  223. 

Five  specimens  were  obtained  in  the  Anseba  valley 
between  4,000  and  5,000  feet  above  the  sea.     This  1 
Tortoise  chiefly  inhabits  the  thickets  on  the  banks  of 
river,  and  under  bushes  in  the  neighbourhood.     I  ne 
saw  it  on  open  ground. 

2.  Pelomedusa  gehafie,  EUpp. 

Gray,  Cat  Tort  Croc.  &c.  Brit  Mas.  p.  38.;  Gat  Shield  B 

p.  53. 
Pentonyx  gehafie,  Biipp.  Neo.  Wirb.  p.  2.  t  L 
Pdomediisa  gehafias,  Gray,  Cat  Tort.  Croc  &c.  Brit  Mus.  p. 

Cat  Shield  Bept  p.  63. 

Not  uncommon  in  the  Anseba  and  its  feeders.  I  a 
found  it  in  the  Lebka.  During  the  dry  season  th< 
rivers  completely  disappear,  and  this  Tortoise  must  hv 
itself.     It  has  a  most  ofiensive  smelL 

^  For  the  determinations  of  the  minority  of  the  reptiles,  I  am  indebtec 
the  kindness  of  Dr.  Giinther. 


REFTILLt,  445 

The  name  gehafie  is  taken  from  the  Tigr^an  language. 
To  alter  it  into  gehaficBy  as  Gray  has  done,  is  of  course  a 
mistake. 

Order  SAURIA. 

3.  Chamselio  Isevigatus  (?),  Gray. 

Gray,  Proc.  ZooL  Soc  1863,  p.  96, 

I  obtained  a  single  young  specimen  in  the  Anseba 
valley.  I  am  far  from  certain  that  it  belongs  to  the 
above-named  species,  as,  despite  its  not  being  adult,  the 
crest  is  more  strongly  developed  than  in  the  full-grown 
specimen  in  the  British  Museum.  The  colouring  is  the 
usual  greenish  grey,  varying  according  to  circumstances, 
with  two  rows  of  subquadrate  white  spots  along  the  sides. 

4.  Varanus  ocellatus,  Eiipp. 

Biipp.  Atlas,  t.  vL 

Dum.  et  Bibr.  Erp.  Gen,  iiL  p.  496. 

Regenia  ocellata,  Gray,  Cat  Lizards,  Brit,  Mus.  p.  9. 

One  specimen  procured  in  the  Anseba  valley ;  it  was 
under  a  rock  in  rather  open  ground.  The  spots  are 
scarcely  perceptible,  and  the  head  is  longer  and  not 
rounded  above,  as  in  Eiippell's  figure,  but  there  is  no 
essential  difierence.  The  measurements  are  the  follow- 
ing:— 

ft.    In. 

Whole  length  from  nose  to  end  of  tail 2  7*0 

Length  of  tail  from  anus 1  2*0 

of  head,  about 0  3'0 

from  ear  to  point  of  nose 0  2*75 

of  foreleg  from  shoulder  to  end  of  middle  toe  0  4*75 

of  middle  toe  of  same  and  claw 0  1*25 

of  hindleg  to  end  of  second  toe 0  6*0 

of  longest  toe  (second)  and  claw 0  1*3 


446  ZOOLOGY. 

5.  Agama  (Siellio)  cyanogaster,  Riipp. 

Riipp.  Neu.  Wirb.  Kept.  p.  10,  t.  t.— Dam.  et  Bibr.  ir.  p^  &32. 

Extremely  abundajit  throughout  the  highlands,  where 
may  often  be  seen  upon  rocks.  It  is  especially  comnu 
in  Northern  Tigrd  on  sandstone.  It  is  a  large,  bright] 
coloiired,  and  handsome  lizard,  and  very  active  and  war 
so  much  so  that  I  could  only  obtain  specimens  by  shoo 
ing  them.  I  never  saw  it  in  the  Anseba  valley,  nor  i 
general  below  7,000  feet  of  elevation. 

The  blue  veining  of  the  under-parts  is  far  more  coi 
spicuons  in  some  specimens  than  in  others ;  usually  it 
to  a  great  extent  confined  to  the  chin.  The  followir 
are  the  dimensions  of  a  moderate-sized  specimen : — 

Whole  length 10 

Length  of  head 1-2B 

„      of  tail  from  anUB ft-7 

„      from  ear  to  end  of  now 1*0 

„      of  foieleg  with  toe   ...    ' 8'3 

„      of  foorth  toe  of  forefoot 0-6 

„      of  ilindleg  to  end  of  lougeet  toe Si) 

„       of  fourth  toe  of  hindfoot 0-6 

6.  Agama  annectans,  sp,  nov. 

A.  T^fo-fatea,  nigro-marmorala,  tquamii  omnilnu  partnt,  fere  aqtuUHn 
eapitt  d^pTtuo  tublriangulari,  re^toniiui  parolieit  tt  inteneajmUo  ^larn 
ipinosit,  dorio  non  criitato,  eauda  dongala,  otteMiala,  iqwunU  caudal^ 
antiee  dutiiieU  atmuiaUi,  poitic6  rubimbrieatii. 

1  1 

Long,  tota 12-0  — 

„      eapUii 1'3  1-6 

„     eorporU SS  4-0 

„      cavda  ab  ano 7'5  — 

„     pedis  anlerion$  cum  dxgitit     ...      2'8  3D 

„      ejiudem  dig.  primi  cum  urufni     .     .       0'4  0"4 

„  „  Hcundi       „  .    .      0-47  0'5 

„  „  (ertii  „  .    .      0-6  0-9 


¥ 


REPTIUA. 

447 

Long,  ^uidem,  dig,  quarti  cum  latgui 

.      0-6 

i 
0-66 

„             „          guinti        „ 

0-62 

0-56 

„     ftdit  poderiorii  cum  dig.    .     . 

3b 

3-9 

„     ejusdtm  dig.  primi  cum  ungni 

0-36 

0-4 

$€cundi       „ 

0-55 

0-6 

„               „           UHii           „ 

076 

0-8 

quarli         „         . 

0-8 

0-87 

quiiiti         „         . 

0-7 

07B 

Bab.  howl  procnl  a  lillore  maTi$  Erythrtti  in 

■irea  2,000  ped. 

Rufous  brown  (in  spirit)  with  irregular  blackish  rings, 
and  an  indistinct  whitish  line  down  the  centre  of  the 
back.  Scales  small  throughout,  and  subequal.  Head 
Bubtriangular,  flat,  the  length  twice  the  height  at  the 
eyes,  and  exceeding  the  breadth  in  the  proportion  of 
4   to   3.    Nose  rather  blunt,  terminated  by  a  single 


^ 


I 


448  zooLOor. 

puDctulated  rostral  scale,  much  more  broad  than  high;  □< 
trils  towards  the  hinder  part  of  a  swollen  subovate  sea 
separated  by  one  or  two  scales  from  the  rostral.  Canth 
i^straMs  and  supra-orbital  ridge  prominent,  terminatii 
at  the  nasal  scale.  Frenal  region  depressed,  covered  wi 
small  elongate  scales.  Forepart  of  the  head  covered  wi 
nearly  flat  scales,  those  in  the  centre  a  little  longer  thj 
the  others ;  occiput  covered  with  flat  scales  towards  t 
forehead,  whilst  behind  and  towards  the  temporal  regio: 
they  are  slightly  keeled.  Ear-opening  large,  nearly  equ 
to  the  eye.  A  few  spines  scattered  aronnd  it,  especial 
behind  eyelid,  covered  with  small  scales.  Mental  ahiei 
almost  oval,  a  little  broader  than  high.  Supnt-labiala  aboi 
14,  not  extending  to  the  gape,  the  row  of  larger  scali 
constituting  them  being,  however,  continued  backward 
and  terminating  in  a  smaU  spine  above  the  gape.  Infr: 
labials  12 — 13,  also  not  extending  back  quite  as  far  s 
the  gape.  Chin  covered  with  small  rhomboidal  scale 
decreasing  in  size  backwards.  Throat  with  two  or  thie 
irregular  cross  folds.  Back  not  crested,  covered  witi 
small  imbricate  scales,  those  near  the  centre  larger  an« 
slightly  keeled,  especially  towards  the  fixint.  A  fe^ 
scattered  spines  of  small  size  between  the  shoulders 
TaU  longer  than  the  head  and  body;  scales  a  littlt 
larger  than  elsewhere,  strongly  keeled,  disposed  in  narrow 
rings  in  front,  but  becoming  gradually  imbricate  towardi 
the  tip.  Scales  of  the  hcWy  small,  smooth,  rhombic,  sub 
equal.  Pre-anal  pores  eight  in  a  single  row.  Third  and 
fourth  toes  in  the  fore  foot  subcqual ;  in  the  hind  foot  th( 
third  toe  is  veiy  little  shorter  tlian  the  fourth.     Point  ol 


liEPTIIJA.  449 

longest  hind  toe  reaxjhing  the  eye  when  the  hind  leg  is 
laid  along  the  side.  Fore  toe  just  extending  back  to 
the  thigh. 

I  obtained  two  specimens  of  thiB  apparently  new 
Agama  near  the  base  of  the  pass  leading  fix>m  Komayli 
to  the  highlands.  In  one  the  tail  is  imperfect.  The 
species  forms  an  immiatakeable  link  between  Agama 
and  StelliOy  for  although  the  structure  of  the  tail  scales 
is  that  of  the  latter,  they  being  arranged  in  distinct 
rings,  the  animal  otherwise  agrees  best  with  the  former. 

7.  Agama  colonorum,  Daud. 

Lacerta  ctgapia,  L.  Syst.  Nat.  ed,  12,  voL  L  p.  367. 
A'  colonorum^  Paadin,  Hist  Bep.  iiL  p.  366.-^Dam.  et  Bibr.  iv.  p. 
489, — Riipp.  Neu.  Wirb.  p.  14,  t.  iv. 

This  Lizard  was  common  on  granite  rocks  about  Eairo 
near  Af  Abed,  north  of  the  Lebka,  at  about  3,000  feet 
above  the  sea.  The  head  in  many  specimens  was  bright 
blue  in  colour,  chin  scarlet.  Doubtless  these  colours  are 
seasonal  as  in  Calotes. 

8.  A.  occipitalis,  Gray. 

Gray,  PhiL  Mag.  1827,  p.  264. 

A.  coUmorumf  var.  Dum.  et  Bibr.  Erp.  Gen.  iv.  p.  400. 

A  single  specimen  taken  at  Antalo  on  the  highlands 
with  Stellio  cyanogaster. 

9.  Lacerta  samharica,  sp.  nov. 

Dam.  et  Bibr.  y.  p.  278. 

L,  ffriaea,  9upeme  lonffiiudinaliter  fusco-fcueiata  et  maetUaiostHgata, 
sqtuimis  dorMlibui  anUse  rotundatis  e<mvex%8,  poHice  carinatis,  ventralibui 
Q-seriatiSy  rhombeis^  preanaU  una  majori  ;  squamis  postoeeipitalUmi  magnii 
a  minoribus  polygonis  eircwndoHs,  naribuB  in  angulo  poiteriori  et  inferiori 
squama  inflatoi  ncualis  poiitis^  et  a  iquama  na§aH  duobue  pott  wuaUhui  et 

0  G 


450  nooioor. 

prima   lupraltibiali   (ireumdati*;   eauda   ditplo  qwim   eorpore  earn  ea} 
longiori, pede potttriori  ad  humtrum  attingenle. 

I  a 

fA>ng.tola ■     ■  1^75  12-25 

„     eapitU 0-95  0-85 

Lot.  tjuidem 0-&2  0-« 

Long,  eorparia 3-4  2'98 

„      tauda  abano 876  9-0 

.,     pedu  aHterwrit  enm  digitii  .    .    .  1'26  I't 

„     gtud^mdigUipnmiatinmti/ui  O'S  0-18 

„               „            Mcunrft             „  0-35  (V25 

„               „            lertiittquarli  „  04  0-36 

„                 „             quinti                „  0-3  0-84 

„     ptdit  poiUriorit  euiit  digUii  ...  S'S  S'15 

„      rjutdim  digitit primi aim  ungni    .  022  0*22 

„               „             itexmdi        „  035  OM 

„                 „              Itrlii             „  0-6  0-46 

„               ,,             quarii         ^  08  07B 

„                                 qiiinli          „  0-4  0'35 

Colour  grey,  with  varying  diisky  lines  and  rows  < 
spots  aloDg  the  back.  Head  not  quite  twice  as  long  i 
it  is  broad,  bluntly  conoidal ;  nostrils  in  the  lower  hind' 
angle  of  the  nasal  shield  juBt  above  the  first  upp 
labial,  with  two  post-nasals,  the  upper  of  which,  howevt 
scarcely  reaches  the  nostrlL  Two  loreal  scales,  one  befo 
the  other,  the  posterior  largest.  A  large  poet-occipit 
shield  on  each  side  surrounded  by  snaaller  polygon 
shields ;  between  the  two  is  an  elongated  pentagon 
shield,  with  a  small  ring-shaped  depression  in  the  centi 
one  shield  behind  it  and  two  in  front.  Two  supra-orbit 
shields  on  each  side,  separated  from  the  supra-orbit 
ridge  of  small  shields  by  a  row  of  small  granular  acalt 
Lower  eyelid  covered  with  small  granular  scales.  Upp 
labials  about  10,  the  seventh  from  the  rostral  expand 
alwve,   and   forming  the  lower  portion  of    the   orb 


RSPTIIIA.  451 

Lower  labials  about  7.  Collar  distinct  of  about  eight 
enlai^ed  scales.  Ventral  shields  rhombic  in  six  rows, 
a  row  of  smaller  semicircular  shields  along  each  side 
of  them.  Scales  of  the  back  rounded  in  front,  be- 
coming rather  larger  and  keeled  behind,  thoao  of  the 
tail  strongly  keeled,  arranged  in  distinct  rings.   Fre-anal 


shield  of  moderate  size,  with  about  three  rows  of  smaller 
shields  between  it  and  the  thighs.  About  10  to  11 
femoral  pores  on  each  thigh.  Hind-foot  laid  forward 
reaching  to  the  shoulder,  fore-foot  laid  back  reaching 
about  half-way  to  the  thigh.  Tail  compressed,  about 
twice  as  long  as  the  head  and  body. 

This  Lizard  was  common  near  the  coast.     I  mifortu- 
nately  only  brought  two  specimena  away,  however. 


453  zooioor. 

lo.  L..  Sturti,  sp.  aov. 

L.  perafittii  praadtnti,  $»d  aauda  Imtgiori  magi*  rottMdalo,  dig 
br»mf»ib%M,  igiMiina  pn-anaii  mtyori  rmiia,  difitw  quartvs  pedit  po*ttri< 
(HirmattmgiL 

FoQ.  Ai«L 

"    Long,  tola 13-0 

„      eapUii 0-75 

Lot.  cjiudmt 0'46 

Long,  eorporit S'4& 

„     cawias  ab  ano 9*8 

„     pedi*  anteriorii  e.  digiti* -  I'l 

„     gi(«d«m  digiii  primi  e.  mtgui 0*15 

„  „  MCUnd*     „         0'3 

„  „  ttrtU  et  qvarti  c.  %ngtii  ....  0*3li 

„  „  qviiUi  „       ....  0*28 

„     ptdit  poiUnorit I'86 

„     ^wtUm  digUi  frmi  e.  ung 0-SS 

„  „  McuWi    n        0-38 

„  „  Urtii         0-43 

„  „  giiarM      „        0-67 

„  „  quint*      „        0*4 


But  for  the  difierence  in  the  pre-anal  shields  I  shoi 
not  describe  this  specimen  aa  a  separate  apet^es  from  t 
last,  the  other  distinctions  being  very  trifling.  The  he 
is  a  little  shorter  and  blunter,  but  the  shields  are  v€ 
similar,  there  being  a  very  small  additional  shield  1 
tweea  the  two  post-fr(»ital3.  Supra-labiala  9,  the  mi 
from  the  rostra]  expanded  above ;  infra-labials  6.  Shiel 
of  the  collar  about  14,  but  they  pass  so  completely  ii 
the  smaller  scales  at  each  end  that  it  is  difficult 
assign  a  precise  number.  Ventral  shields  in  six  rows,  w: 
a  row  of  smaller  flat  ^elds  at  each  side  of  the 
Scales  of  back  granular,  passing  gradually  behind  ii 
the  strongly-keeled  scales  of  the  taiL     Toes  compress 


REPTIMA. 


453 


not  keeled  beneath.  Hind-toe  reaching  forward  to  the 
ear,  fore-toe  extending  back  more  dian  two-thirda  of  the 
distance  to  the  thigh.  Femoral  pores  12  in  each  thigh. 
Pre-anal  shields  all  smaller  than  the  ventrdl. 


A  single  specimen  was  obtained  by  Captain  Bturt,  to 
whom  I  am  indebted  for  it,  and  after  whom  I  have 
named  it,  in  the  Komayli  pass,  near  Undul  Wells. 


II.  Acanthodactylus  mucronatus,  ap.  nav. 

.  A.  ffriteo-fulva,  fascid  mtdiand  /u»ed  dortUi,  ftrifUjiM  laltrcdiiiu  tx 
griuo  tt  fiuco  compotitit  omata,  pedibu*  rupra  /lueit,  grUa}-vuu:iilat'U  ; 
nnri  in  medio  i  »quamanim  potitd  duarum  aateriorum,  duanaa  poiUriorum, 
$qiuimii  BJipra-orbitalHnu  utrinque  diiohas,  granvlU  siij>ra  infra  tt  poatice 
tetl  non  antif*  cireuindalit,  vertieali.  Eloiigato-penlagoHali,  in  medio  milcaid, 
po*l-oeeipit(tlibut  magai*,  polygonU  ignamit  parti*  4   valde    iiurqttaliimi 


454 


ZOOLOGY. 


mttrvaatntOnu.   Pedit  potteriorii  difiiiu  fert  ad  roftrtim  aitingit.     Sp 
pre-an(Uu  magna,  ntinortbiu  cirtumdata. 

Long,  tola 6.5 

„     eapitit 0-68 

Lot.  ejtutfem 0^ 

Long.  corporU 1  '46 

eavda  ah  aiio 4*6 

ptdit  anUriori* 078 

tytudem  digiti  tertii  {el  juart*)  e.  laigtii     .    .    .  O'SS 

ptdii  potterioru I'M 

gutdvm  digiti  ftimi  e.  wigui O'SS 

„  teeundi O'SS 

„  (ertM         , 0-36 

„  quarti       „        0-63 

„  quinii       „        0-36 

Greyish  tawny  in  colour,  with  a  dai^  band  along 
centre  of  the  back,  and  others  of  dusky  mixed  w 


grey  at  the  sides.  Legs  dusky  above,  spotted  pi 
Snout  very  pointed,  somewhat  alternate ;  nostrils  in  ) 
centre  of  four  small  shields,  two  anterior  and  two  p 
terior,  a  distinct  canthus  rostralis  passing  from  1 
supra-orbital  ridge  to  the  nostrils ;  frenal  region  fl 
vertical,  consisting  of  two  shields,  tiie  posterior  by 


REFTILIJ.  455 

the  largest.  Two  large  supra-orbital  shields  ou  each  side, 
the  two  surrounded,  except  in  front,  by  small  granular 
scales ;  between  them  in  front  is  an  elongately  pentagonal 
vertical  shield,  ^vith  a  central  depressed  groove ;  post- 
occipital  shields  large,  separated  by  a  row  of  small  shields 
of  very  unequal  size.  Lower  eyelid  covered  with  small 
granular  scales.  Upper  labials  8  or  9 ;  of  which  six 
or  seven  are  of  moderate  size,  then  a  large  sub-orbital 
shield,  expanded  above,  and  forming  the  lower  part  of  the 
orbit.  Lower  labials  6 ;  sub-mental  shields  4  on  each  side, 
the  hindmost  about  three-quarters  the  length  of  the  other 
three  together.  Collar  not  very  distinct,  composed  of  about 
six  larger  scales,  the  central  being  the  largest,  passing  at  the 
extremities  into  the  ordinary  neck  scales.  Large  ventral 
shields  in  six  rows,  with  a  smaller  row  at  each  side.  A 
large  pre-anal  shield,  surrounded  by  a  ring  of  smaller 
ones.  Femoral  pores  18  in  each  thigh.  Dorsal  scales 
small,  granular.  Caudal  scales  oblong,  strongly  keeled. 
Hind  legs  long,  the  toe  when  laid  forward  almost  reaches 
the  end  of  the  nose,  the  fore-foot  laid  back  reaches  three- 
quarters  of  the  distance  to  the  thigh. 

A  single  individual  wais  procured  in  the  Anseba  valley, 
where  this  lizard  apparently  abounded.  Unfortunately 
no  other  specimens  were  taken,  as  I  had  confounded  it 
with  the  next  species.  It  closely  allied  to  Eremias 
Brenneriy  Peters  (Monatsber.  K.  Preuss.  Acad.  1869, 
p.  432) ;  but  diflfers  in  several  points,  especially  in  the 
number  of  supra-labial  shields,  the  proportion  of  the  sub- 
mental shields,  and  the  greater  length  of  the  limbs. 
The  colour  also  is  very  different. 


456  EooLoer.    , 

13.  A.  boskionus,  nu. 

Dum.  et  Bibr.  t.  p.  S7a 

Common  in  Samhar.  Two  Bpecimens  coUected  ne 
Zulla. 

13.  Eupreasis  quinquetseniatus,  Licht. 

ScinoM  fuwtgtictonMUtu,  lieht.  V«k.  DonbL  Miu.  Bori.  ph  103. 
£t)preuH  Savignii,  Dum.  «t  Bibr.  Erp.  G«ii.  t.  p.  677. 

1  obtained  one  large  specimen  at  Sum,  and  foi 
smaller  ones  on  the  highlands,  three  of  them  from  tl 
stomach  of  a  Secretary-bird.  They  differ  greatly  i 
markings,  but  Dr.  GUnther  considers  all  to  be  rarietit 
of  the  same. 

14.  E.  Perrottetii,  Dum.  et  Bib. 

Dum.  et  Bibr,  Brp.  Gen.  t.  p.  669. 

?  Tiliqua  Bmioni,  Blyth,  Jour.  M.  Soc.  Benj;. 

This  was  obtained  by  Mr.  Jease  at  Wonber  Harattib,  J 
t^e  Lebka  valley.  I  did  not  procure  a  specimen.  I  em 
pect  that  Tiligua  Burtoni  of  Blyth  from  Somali-lane 
the  type  specimen  of  which  in  the  Calcutta  Museui 
has  unfortunately  been  lost,  may  very  possibly  be  tli 
species. 

15.  Gongylus  ocellatus,  Forsk. 

LoMiia  ocetiata,  Fonk.  Desa  Anim.  p.  13. 
0.  oceUatut,  Dum.  et  Bibr.  Erp.  Gen.  v,  p.  616. 

Several  specimens,  taken,  with  Eupressis  guingtteia 
niatus,  from  the  stomach  of  a  Secretary  bird  shot  a 
Aehangi.     All  are  small. 


REPTIUA.  457 

Order  OPHIDIA. 

z6.  Typhlops  Eschrichtii,  SchlegeL 

Schleg.  Abbild.  AmpL  p.  37,  ^  32,  figs.  13—16. 
OphikalnMum  EkkrHclUii,  Ihun.  et  Bibr.  vL  p.  266. 

I  am  indebted  for  a  single  specimen  taken  at  Senaf^ 
to  Capt.  Stnrt.  It  is  small^  measuring  only  12  in.  in 
length. 

17.  Echis  arenicola,  Boie. 

IsiB,  1827,  p.  556. 

E,  variay  Reoss.  Mus.  Senkenb.  iL  p.  160,  t.  viL  £  2. 

This  viperine  Snake  was  common  in  the  camp  at  Zulla, 
individuals  frequently  being  found  about  the  tents,  and 
it  is  surprising  that  no  one  suffered  from  its  bite.  All 
the  specimens  seen  by  me  were  smalL  I  preserved  two, 
one  of  which  has  193  ventral  shields. 

18.  Dasypeltis  abyssinicus,  Dum.  et  Bib. 

Bachiodon  ahyssinus,  Dumeril  et   Bibron,  Erp.  €kn.  viL  p.  496, 
pL  81,  figs.  1  and  2. 

A  single  specimen,  quite  young,  only  12  in.  long,  was 
obtained  at  Senafd  by  Lieut.  Sturt,  to  whom  I  am  in- 
debted for  it.  Dr.  Gunther  informs  me  that  he  doubts  if 
this  species  be  distinct  from  D.  palmaruin.  The  tail 
measures  If  in.  The  colours  are  not  so  bright  as  in 
Dumeril  and  Bibron's  figure. 

19.  Boodon  lemniscatum,  Dum.  et  Bibr. 

Dumeril  et  Bibron,  Erp.  Geu.  vu.  p.  365. 

I  obtained  two  specimens,  one  at  Ashangi,  the  other  at 
Adabagi.  The  former,  a  male,  measures  21  in.,  the  latter, 
a  female,  30  in.,  of  which  the  tail  is  4in.  The  male  is 
much  the  darker  in  colour. 


458  ZOOLOGY. 

20.  Homalosoma  lutriz,  L. 

CoIuW  Ivtrix,  L.  S;Bt  Nat.  ed.  IS,  L  p.  879. 
Homalotoma  hUrix,  Dum.  et  Kbr.  vii.  p.  IKK 

A  very  young  specimen,  only  9  in.  long,  was  takei 
ABliangi. 

21.  Bucephalus  typus,  Smith. 

Zoologioil  Jomnal,  vii  p.  441. 

B.  eapmtii.  Smith,  IlL  ZooL  Soutli  Alt.  RepL  pL  IS.— Giintlwi 

B.  lyput,  Dom.  et  Bibr.  Erp.  Gen.  viL  p.  877. 

One  specimen,  only  22  in.  lon^  was  procured 
Senaf^ ;  a  larger  one,  which  was  captured  on  the  Aasc 
measures  33  in. 


32.  Psammophis  sibilans,  L. 

Cobiher  nbOani,  L.  gyat.  Nat  ed.  IS,  p.  383. 
Phammcphu  monoligtr,  Dum.  et  Bibr.  vii.  p.  891. 

A  single  individual  of  this  common  African  Tree  Sn; 
was  taken  in  the  Anseba  valley.     It  is  23  in.  long. 


Ohdkr  BATRAOHIA. 

33.  Pyzicephalus  rugosus, 

A  single  specimen  found  at  Scnaf^. 


34.  Rana  mascariensis,  Dam.  et  Bibr, 
Dum.  et  Bibr,  Eip.  Qea.  viii.  p.  350, 

This  Frog  was  common  in  the  streams  at  Suni,  anc 
Ain  on  the  Lebka,  both  at  low  elevations,  not  exceed 
2,000  feet  above  the  sea. 


REFTILIA.  459 

25.  Dactylethra  Isevis,  Daud. 

Bufo  leviSf  Daud.  Hist.  Nat.  Baiu.  p.  85,  pi.  30,  f.  1. ;  Hist.  Rept. 

viiL  p.  171. 
DactyleUira  eapenns^  Dmn.  et  Bibr.  viiL  p.  765. 

I  obtained  two  specimens  at  Senafd. 

26.  Bufo  pantherinus,  Boie. 

Dam.  et  Bibr.  Erp.  QeD.  viiL  p.  687.  ' 

B.  ardbicuSf  Riipp.  Atlas,  t.  iii.  f.  2. 

B,  regulariSf  Beuss.  Mus.  Senck.  i.  p.  60. 

This  Toad  abounded  about  Lake  Ashangi  and  other 
places  on  the  highlands.  It  is  highly  variable;  some 
specimens  being  covered  with  small  tubercles,  others  quite 
smooth.  I  also  obtained  some  young  specimens,  appar- 
ently belonging  to  the  same  species,  at  Ain  in  the  Lebka 
valley,  only  about  1,200  feet  above  the  sea. 

27.  Bufo,  sp.  indet. 

A  single  young  Toad  captured  at  Sum  is  too  im- 
mature for  determination.  It  appears  to  belong  to  a 
distinct  species  fipom  the  last. 


Class  PISCES. 

I  obtained  but  two  species  of  Fish.  The  route  taken 
by  the  army  avoided  all  rivers,  and  in  Lake  Aahaugi  1 
was  unable  to  capture  anything,  and  only  procured  one 
specimen  from  the  stomach  of  a  Grebe.  The  fish  taken 
are  both  Cyprinoids ;  I  am  indebted  to  Dr.  GUnther  for 
their  names. 

I.  Capoeta  DilloAit,  Got.  et  VaL 

I  preserved  a  skin  of  a  Fish  taken  in  a  stream  at 
Agula,  which  Dr.  GUnther  is  inclined  to  refer  to  this 
species,  although  it  difiers  slightly  &om  the  description. 
The  dorsal  fin  contains  3  spiny  and  9  soft  rays,  whereas 
C  Dillonii  is  said  to  have  only  1  spine.  There  are  32 
scales  on  the  lateral  line. 

a.  Discognathus  lamta,  Hans  Bach. 

I  obtained  several  specimens  in  the  stream  at  Sum, 
and  others  were  caught  by  Mr.  Jesse  in  a  small  pool  left 
in  a  watercourse  at  Amba,  about  twenty-five  miles  north 
of  Maasowa.  These  differ  in  no  respect  &om  specimens 
from  Palestine  in' the  British  Museum.  They  aU  have 
the  black  mark  at  the  base  of  the  gill  opening,  although 
the  lateral  marks  are  less  distinct  than  in  Indian  speci- 
mens. The  scales  on  the  lateral  line  are  33  to  34  in 
number. 


PISCES.  461 

The  same  Fish  has  been  obtained  in  abundance  by 
Lieut  Mockler  at  Lahej,  near  Aden,  in  a  small  stream 
which  occurs  there. 

Singular  to  state,  the  single  Fish  procured  at  Lake 
Ashangi,  though  not  in  sufficiently  good  condition  for 
accurate  determination,  appears  also  to  belong  to  this 
species.  It  is  of  course  possible  that  it  may  be  a 
distinct  form  occurring  on  the  highlands,  but  it  is 
clearly  very  closely  allied,  if  not  identical 


Ilk. 

I 

i 


MOLLUSCA. 


OCEANIC  MOLLUSCA  OBTAINED  OFF  THB  S.B.  COA 
OF  ARABIA. 


Class  PTEROPODA. 

1.  lIgal<Ba  wtcinata,  Bang. 

2.  H.  Itmffiroitrit,   Leanenr. 

3.  ^rialia  troch^ormit,  iJ'Orb. 


Scarce. 

Abimdant. 

Rare. 

4.  Styliola  comi/ormii,  D'Orb. 

Not  abundant 

Class  GASTEROPODA. 
D.  Litiopa  bomhyx,  Kang. 
Not  very  common  :  attached  to  floating  weed. 

Class  HETEROPODA. 

6.  Janlhina  fragilit.  Lam.  (not  Heevo). 
Syn.  /.  roteota,  £t.  and  /,  affinii,  Bv. 

Common :  the  variety  is   that  called  /.  roseola 
Reeve,  but  it  appears  to  be  merely  a  variety  of  fragi, 


MOLLUSC  J.  463 

7.  /.  globosa,  Swains. 

Not  rare :  a  variety  with  rather  a  high  spire. 

8.  /.  hijida,  NuttalL 

Less  common  than  the  other  two,  but  not  rare.     It  is 
near  /.  Vinsoni,  Deshayes. 

9.  A  tlanta  Peronii,  Lesueur  and  Blainv. 


11. 
MARINE  MOLLUSC  A  FROM  ANNESLEY  BAY. 

Class  GASTEROPODA. 
Obdkh  PROSOBRANCHIATA. 

Family  MTJBICIDiE. 

1.  Murex  anguliferus,  Lam. 

2.  M,  ternispina^  Lam. 

3.  M.  scolopaXf  Dillwyn. 

4.  M.  margariHcola,  Brod. 

5.  Tritonidea  rubiginasa,  Bv. 

A  smaU  variety,  with  fine  sculpture. 

6.  Hemifusxu  paradinacus,  Chem. 

Family  TBITONIDiE. 

7.  Persona  cancelUna^  Desh. 

8.  Ranella  tubercuUUcLj  Brod. 
0.  R,  spinoia,  Lam. 


Very  common. 


Family  BUCCINIDiE. 
10.  Noisa  pullus,  L. 


464  ZOOLOOT. 

11.  K.  KUneri. 

12.  N.  plieaUUa,  A.  Ad.  var. 

The  sculpture  a  little  more  nodose,  and  the  f< 
rather  lesa  tumid  than  in  the  types  from  Mr.  Cumii 
cabinet,  now  in  the  British  Museum. 

13.  31  naryimtlata  (I)  Lam. 

A  very  amall  form,  approaching  H.  delicata,  Rv. 

Famu-t  PURPTJRIDiE. 
14.  Purpura  lUppocattanum,  L, 
The  only  species   found.     None  of   the  usual  ty 
allied  to  P.  persica  and  P.  carinata  were  observed. 

15.  Sutrum  daUim,  Knum 

16.  S.  anaxartt,   Dncloo. 


Very  abundant. 
Rare. 


Familt  OLIVID^ 
17.  Oliva  infiata,  Iaio. 


18,  Ancillaria  exiffua,  Sow. 

Fahilt  Mrniisie. 

19.  Mitra  aitud/ilii,  Br. 

20.  M.  prdiMa,  Rv. 

21.  Columbella  variant,  Bow. 

22.  O.flava,  Bnig. 

23.  J^i^na  mendieana.  Laid. 

Fawlt  MARGINELLro.<E. 
S4.  MargifteJia  monilit,  L. 

Family  NATICID-^ 

25.  Natka  mammilla,  L. 

26.  X.  melanoetoTna,  Gm. 


MOLLUSC  A.  465 

Family  EULIMIDiE. 

27.  Leiostracci,  sp.  nov. 

This  is  closely  allied  to  L.  pura,  A.  Ad.,  but  it  has  fewer 
whorls,  and  the  mouth  is  differently  shaped,  the  columella 
being  straight  instead  of  sinuate.  The  specimens  were 
found  parasitic  on  a  species  of  ScutelkL 

Family  SOLARIID^ 

28.  Solarium  penpectivum^  L. 
Syn.  S,  nobUe,  Hinds. 

Family  CONLDiE. 

29.  Conits  arenatu9f  Hwass. 

30.  C,  Adanscnif  Lam. 

Family  TEREBRIDiE. 

31.  Terebra  einetella  (1),  Desh. 

Only  one  specimen  found,  which  is  in  poor  condition. 
It  belongs  either  to  the  above,  or  a  closely  allied  species, 

32.  T.  nasBoideSy  Hinds* 

The  plaits  are  less  numerous  than  in  the  British 
Museum  specimens,  but  this  character  is  evidently 
variable. 

Family  STKOMBIDiE. 
33.  Strombus  tricomis,  Martini 

Conunon  about  coral  reefs.  Several  living  specimens 
were  washed  up  at  Malkatto  one  day  when  the  sea  was 
rougher  than  usual. 

34.  S.fasciatuSf  Bom. 

35.  S,  floridu9y  Lam, 

36.  S,  elfganty  Sow. 

H  H 


37.  S.  dentatu*.  L. 

38.  Rottellaria  curvirottru,  Iah 

Common  on  sandy  shore  at  Malkatto. 

Familt  CYPILEID^ 

39.  Cffpnea  tardus,  L. 

40.  Trivia  ornza.  Lam. 

Familt  CERITHIID.1E. 

41.  CeritAum  atpertam,  Desh. 

42.  C.  varUffotttm,  Qaoy. 


Very  common. 
Not  rare. 


43.  C.  Adeneme,  Ry. 


44.  C.  inlentriatumt  Sow. 
46.  C.  Koehii,  Pha 

46.  C.  bi/ateiatuntf  Sow. 

47.  C.  sp.  noT. 

A  single  specimen  in  poor  condition. 

48.  C.  eoluvtna,  Sow. 

49.  TympanoUmo*  Layardi,  Ev. 
fiO.  T.fiuviatiU,  Pot.  and  Mich. 
51,  i'ymtw  po/wrfrM,  Brag. 

Common  amongst  mangroves. 

62.  Bitlium,  sp.  nov. 
A  single  specimen. 

53.  Triforii  ruher.'Q.iD.is. 
H.  T.  dittinetwi,  Desh. 
65.  Faaigulla  (I)  Bp. 
A  single  specimen  in  bad  condition. 

Family  CEEITHIOPSID-E. 
5C,  Gm'lMop«i$,  ap. 
Apparently  new. 


Common. 


MOLLUSC  A,  467 

Family  LITTORINIDiE. 

57.  Littorina  malaccana,  Phil. 

58.  L,  intermedia,  Phil. 

Family  PLAXAXIDiE. 

59.  PlanaxU  mlcata.  Born. 

Syn.  P.  unduiata,  Lam, 

Family  RISSOID^ 

60.  Rissoina  myosoroides,  BecL 

61.  E.  plicata,  A.  Ad. 

62.  7?.  medial,  Schwartz,  v.  Mohr. 

63.  B.  d'Orbignyiy  A.  Ad.  var. 

64.  7?.  sp.  nov. 

65.  AUiba,  ftp.  nov. 

Family  NERrTIDiE. 

66.  Nerita  polita,  L. 

Very  common. 

67.  N,  rudit,  Gray. 

Common. 

68.  N,  cdbicetta,  L. 

Common. 

Family  TROCHIDiE. 

69.  Tectus  dent^Uus,  Forsk. 

70.  T,  nodultferus,  Lam. 

71.  EucluHus  edentulus,  A.  Ad. 

72.  B,  baccatus,  Ktister. 

73.  Trochus  crehrUiratrUy  Jonas. 

74.  Monodonta  cancUiferOj  Ijam, 

75.  Gibbula,  sp.  nov. 

Near  G.  adriatica,  Phil.,  but  the  spire  is  higher,  and 
the  sutures  less  impressed.     It  also  approaches  G.  leu- 

H  H  2 


468  zooLoor. 

costicta,  A.  Ad.,   but  the  whorls  are  flatter,    and 
spire  higher. 

76.  Gma,  sp.  hot. 
A  peculiar  carinate  speciea 

Familt  TUEBINID.^ 

77.  TuriM  CfumnUtiamu,  fir. 

78.  T.  ap. 

I  caiLDot  find  any  name  for  this  species ;  it  is  a  ai 
form  figured  in  the  "  Descr.  de  I'Egypte,"  pi.  v.  fig.  2 
7£>.  T.  ooronatut,  Gmel. 


Familj  FISSUEELLID-E. 
80.  FittvrtUa  eya^ulum,  Bt. 


Familt  PATELLID^. 
81.  Patdla  rota,  Chem. 


Common. 

82.  Helcion,  Bp. 
A  single  young  specimen. 


Familt  CHITONID^ 

83.  Chiton  brevi^noiKt,  Br. 

84.  Ch.  ineurvatut,  Carp. 


Obdbb  OPJSTSOBRASCEIATA. 
Familt  BUlLEDiE. 
86.  Bulla  ampulla,  L. 


'MOLLUSCA.  469 

Order  PULMONIFERA. 

Family  AURICULIDiE. 

86.  MelampuB  ccLstanetis,  Muhlf. 

87.  M,  MassauensiSf  Pfr. 

88.  Flecotrema,  sp.  nov. 

It  is  usual  to  consider  the  AuriculidcB  land-shells : 
this,  however,  is  incorrect ;  they  are  no  more  land-shells 
than  Pyrazus  or  Littorina  are ;  the  majority  of  them 
inhabit  estuaries  and  mangrove  swamps,  others  are 
littoral  shells,  like  Patella  or  Purpura. 


Class  LAMELLIBRANCHIATA. 

Family  GASTROCH^XID^. 

89.  Rocellaria  ovata.  Sow. 

A  Mazatlan  species,  but  unmistakeably  identical. 

Family  CORBULIDuE. 

90.  Corhula  crassoj  Ilinds. 

Family  MACTRID^E. 

91.  Mactra  semistdcata,  Rv. 

Family  TELUNID^. 

92.  Asaphis  deflorata,  L. 

93.  JDonax  vittatus,  Lam. 

Syn.  D.  tri/asci-atus,    Ma  we. 

94.  Tellina  cuUer,  Hanley. 

95.  T,  camea,  Bens. 

96.  T.  capsoidesy  Lam. 

97.  StrigUla  Senegalensis,  Hanley. 

98.  Paphia  striata,  Chem. 


vm 


Fahilt  VENEKID^ 

99.  Doginia  hepaiiea,  Sow. 

100.  Chione  nebuloaa,  Lam. 

101.  Ciree  Arabica,  Chein. 

Common. 

102.  C.pectinala,  L. 

103.  C  corrugala,  Chem. 

104.  Rupellaria  maerophgUa,  Desh. 

105.  Pttricola,  sp.  hot. 

Family  CAEDIAD.^. 

106.  Cardium  rugoium,  XiOm. 

Family  CHAMID^ 

107.  Chama  ^ryphoidet,  Phil. 
106.  C.  iottoma,  Conrad. 

Family  LUCINLD^ 
109.  Lucina  excuptrata,  Rv. 

Family  KELLIAD^ 
110.  Kellia,  ap.  sov. 

A  new  and  peculiar  milk-white  sheU. 

Family  ASTARTID.fi. 
111.  CardUa  tuleata,  Bmg. 

Perfectly  identical   with  specimens  from  the 
terranean. 

Family  MYTIUD.fi. 

112.  Sepli/er  Kraasai,  Duuker. 

113.  MoJiola  hepatUa,  Go}Ad. 

114.  M.  sp.  nov, 

A  peculiar  small  species,  with  strong  denticulatio 
119.  Litkodomm  llanltyanuff'K'v. 


MOLLUSCA.  471 

Family  AVICULIDiE. 

116.  Melegrina  margariti/era  (i),  L. 

117.  Avicula  radiata,  Desb. 

118.  A,  sp. 

Perhaps  a  variety  of  the  last. 

119.  Malleus  regula,  YoT9k. 

Family  AKCIDiE. 

120.  Area  imbncata,  Brug. 

121.  A.  (Barbatia)  decussataj  Sow. 

122.  A,  {Scapharca)  incBquivahisy  Brug. 

123.  A»  (Anomalocardia)  scapha,  Chem. 

124.  Fectunculus  roseits  (i),  Rv. 

Family  SrOXDYLIDiE. 
125.  S})ondyluSj  sp. 

I  cannot  discriminate  the  species  of  this  genus. 

126.  Plicatula  irinbricata,  Menke. 

Family  OSTREIDiE. 

127.  0.  mytiloidtSy  Lam. 

128.  0.  sp. 

On  mangroves/ 


III. 

FRESHWATER  MOLLUSCA. 

Of  these  but  very  few  were  obtained,  as  I  had  no 
opportunity  of  visiting  the  large  rivers.  For  a  complete 
account  of  the  species  of  land  and  freshwater  mollusca 
previously  known  to  exist  in  Abyssinia,  see  V.  Marten's 
"  Malakozoologische  Blatter"  for  1865,  voL  xii.,  and 
1866,  voL  xiii. 


472  ZOOLOGY. 

1.  Sythi«ia,  sp.  iiot(1). 

Antalo.   A  single  specimen  only. 

2.  Mdania  ttiber(Mlala,  MUIL 

Stream  at  Ailat,  Samhar,  very  little  above  the  aei 
level 

3.  Lyntua  NaU^etuu,  Enius, 

Stream  at  Gima-Guna,  on  the  highlands.  Som 
specimens  are  precisely  similar  to  those  from  Soutl 
Africa ;  others  are  rather  more  tumid. 

A  larger  variety  was  obtained  in  the  stream  forminj 
the  head  of  the  Haddaa  river,  near  Takonda. 

4.  Phyta  amiorta,  Micfa. 

Lake  Ashangi,  Specimens  from  the  Nile  in  th 
British  Museum  have  a  smaller  mouth.  Others  froi 
Natal  are  very  similar  in  form. 

5.  Phyta,  sp. 

Stream  on  Wadela  plateau,  at  about  9,500  feet  nea 
Yasendy^.  The  two  specimens  obtained  axe  small,  an 
probably  immature.  They  resemble  in  form  P.  foi 
tinalis,  but  are  not  smooth  like  that  species,  nor  ha\ 
they  any  internal  labiation.  They  may  possibly  be  onl 
an  elongated  vaiiety  of  P.  contorta. 

€.  Ancyiutfiuviatilii,  MiiU. 

Streams,  Guna  Guna,  &c.,  Tigr^. 
7.  A.  ep. 

Stream  near  Mai  Wahiz,  Tigr^.  Four  small  specimei 
found.  They  closely  resemble  some  varieties  of  t\ 
Indian  A.  verruca,  Bens. 


MOLLUSCA.  473 

8.  PlanarUs  Buppdli^  Danker. 

Streams  on  highlands  of  Tigr^.  Common.  The  peculiar 
keel  near  the  suture  described  by  Dunker  is  wanting  in 
my  specimens,  and  is  probably  accidental.  The  mouth  is 
very  oblique,  less  so,  however,  than  in  the  Natal  species, 
L.  P/eifferiy  Krauss. 

9.  P.  Bp.  nov. 

Lake  AshangL  Dead  specimens  found  on  the  shore 
with  Physa  contorta.  This  is  very  near  the  Indian 
P.  compressus,  but  the  umbilicus  is  more  open,  and  the 
keel  more  basal  and  less  median.  It  is  very  near 
P.  Natalensis  also,  but  diflFers  in  the  mouth  being  sinuate 
near  the  periphery,  and  less  arcuate  above,  in  the  lower 
surface  being  more  convex,  and  the  sheU  more  com- 
pressed. 

10.  P.  Natalensis  {!)  Krauas. 

Stream  on  Wadela  plateau.  Only  a  single  specimen 
was  obtained,  which  may  be  the  above  species;  the 
whorls  appear  rather  narrower  and  less  ans^ul^te  at  the 
periphery 


IV. 
LAND  MOLLUSC  A. 

Of  these  also  but  a  very  meagre  series  was  obtained, 
the  country  being  evidently  excessively  poor  in  land- 
shells.  No  Cyclostomdcea  have  ever  been  found  in  this 
part  of  Africa,  and  the  large  AchatincB  and  their  allies 
were  entirely  wanting  in  the  coimtry  traversed,  although 


they  appear  to  have  been  found  farther  to  the  soul 
west,  near  Lake  Dembea. 

1.  Htlix  Darnaudi,  F&. 
Common  throughout    Northern   Tigr^,   from    abo 
3,00U  to  8,000  feet  above  the  sea.    I  found  it  at  Und 
Wells,  Senaf^,   Takonda,  Adigrat,  and  in  the   Ansel 
valley. 

2.  ff.  ap.  uov. 

An  ally  of  H.  pisana,  and  still  more  closely  allied 
//.  auJirostrata,  Fer.  It  is  distinguished  from  the  formi 
in  its  much  more  open  umhOicus,  firom  the  latter  in  beii 
smaller  and  leas  solid.  Specimens,  undistiuguishab 
from  those  collected  by  me  in  Abyssinia,  were  brougl 
from  Persia  by  Mr.  Loftus,  and  are  in  the  British  Museu 
collection. 

This  shell  abounded  in  the  limestone  tract  north 
Antalo,  the  aloes  and  some  of  the  bushes  being  ft^uent 
covered  with  it 

3.  n.  eryophila,  Y.  Martens. 

Occasionally  found   in   the  mountainous  country 
Lasta  near  Meshek,  Lat,  &c.    It  has  four  whorls,  and  tl 
last  appears  to  be  rather  less  broad  than  would  appe 
to  be  the  case  with  the  types  described  by  Von  Marten 
but  there  is  no  essential  difference, 

A.  single  worn  specimen  from  Adigrat  is  higher 
proportion  to  the  diameter,  and  may  belong  to  a  distin 
species. 

4.  H.  rivulaTit(\)  Krauss. 

Agula,  Tigr^.   This  agrees  very  well,  except  that  it 


MOLLUSCA.  475 

very   finely  costulate   throughout  rather  than  striated. 
The  measurements  are  identical. 

5.  Zonites,  sp. 

Meshek.  A  single  specimen  only.  It  is  very  close  to 
//.  egenula,  Mor.,  from  Senegal. 

6.   Vitrina  BUppelltana,  Pfr. 
Syn.  r.  Damaudiy  Pfr. 

Common  in  Northern  Tigr^  about  Takonda,  Senaf^, 
and  Adigrat,  above  6,000  feet.  V.  Darnaudi  appears 
to  be  only  a  variety  of  V.  RUppelliana.  I  obtained  both, 
and  intermediate  forms  as  well. 

7,  V.  Abj/8sinica,  Pfr. 

Takonda.  I  only  found  this  shell  once.  My  speci- 
mens are  a  little  more  angulate  at  the  periphery  than 
the  type. 

8.  v.  sp.  nov. 

Ashangi.  Found  alive  abundantly  amongst  long  grass. 
The  animal  is  a  true  Vitrina  and  not  a  Helicanon^ 
having  no  mucus-pore  at  the  end  of  the  tail 

Twa  or  three  specimens  found  at  Lat  and  Antalo 
differ  somewhat  from  this  form,  but  may  be  varieties 
only. 

9.  Succinea  d^hUts,  Mor. 
9  S,  F/eiferi^  ex  Abyssinia,  auci 

Ashangi  There  are  larger  specimens  in  the  British 
Museum  of  the  same  shell,  also  from  Abyssinia,  labelled 
S.  Pfeifferi.  It  differs,  however,  from  all  European  forms 
of  that  species  by  its  much  smaller  spire  and  larger 
mouth. 


476  ZOOLOGY. 

10.  S.  amphibia,  Dnp.  Tar. 
Adigrat  and  Antalo.     The  mouth  is  a  little  roundi 
than  in  EuropeaD  specimenB,  but  there  appears  to  be  i 
important  distiuctioD. 

11.  Bulimut  Abyuinicit*,  Kiipp. 
Hadoda  near  Zulla,  and  Keren  in  Bogos.     I  found  be 
a  single  specimen  in  each  place. 

12.  B.  OUvieri,  Pfr. 

Scarce  in  Tigr6.  Common  further  south  above  6,00 
feet.  I  found  it  first  near  Undul,  but  it  was  scarw 
About  Ashangi  and  throughout  the  Lasta  hills  it  wa 
the  commonest  shell  met  with,  and  it  is  the  finest  ao' 
handsomest  land-shell  found  by  me  in  Abyssinia,  i 
specimen  from  the  Wadela  plateau  is  greatly  elongatec 
There  is  considerable  variation  in  size. 

13.  Stenogyra  gracUit(^)  Hatton. 
Adabagi,  Tigr^.     A  single  immature  specimen  of  tii 
or  some  closely  allied  species. 

14.  pupa  aenopida,  Hutt. 
Syn.  P.  StnnaarmtU,  Pfr. 
P.  SentffoUntit,  Mor. 

Meshek.  I  was  surprised  at  finding  this  commo 
Indian  shell  on  the  Abyssinian  highlands.  The  spec 
mens  were  quite  indistinguishable  from  some  India 
examples,  and  there  can,  I  think,  be  no  question  thi 
the  two  forms  described  by  Pfeifier  and  llorelet  are  tt 


MOLLUSC  A.  477 

15.  P,  sp.  nov. 

Senaf^,  Adigrat,  Agula,  and  Meshek  A  common  form 
in  Tigr^.  I  can  find  nothing  like  it  described.  The 
nearest  known  form  is  perhaps  P.  doliolum. 

16.  F, /ontqna  (1)  Kranss. 

Same  localities  as  last,  and  very  abundant.  It  has  a 
vertical  groove  behind  the  lip,  and  an  indentation  corre- 
sponding to  the  lower  tooth  of  the  dextral  margin, 
neither  of  which  is  described  in  the  South  African  shell, 
otherwise  it  agrees  admirably. 

17,  P.  sp.  no7. 

Agula,  and  near  Takonda,  Tigrd.  Very  near  P.  urn- 
hilicata,  Drap.,  but  there  is  one  whorl  less  and  the 
aperture  is  rounder,  not  being  pointed  below. 

18.   Vertigo,  sp.  nov. 

Agula,  rare.  A  small  reversed  species,  of  which  I  found 
only  two  specimens. 

19.  Clau^lia  Sennaarensisy  P£r. 

Near  Lat,  south  of  Ashangi.  Only  an  immature 
specimen,  but  apparently  belonging  to  this  species^ 
was  found. 

20.  JSnnsa,  sp.  noT. 

Meshek,  Ashangi,  Lat,  Wandaj  pass.  Nowhere  com- 
mon. ■  This  is  allied  to  E.  C7v/stallum,  Mor.,  but  is  a 
much  longer  and  more  slender  shell,  with  fewer  whorla 


m 

,f  ii'i 


7 


INDEX. 


^ 


INDEX. 


A. 

Abyssinian  houses,  59 ;  Abyssinian 
indignation  meeting  at  Atala,  73; 
scramble  for  the  discarded  stores,  96  ; 
author's  conversation  wiUi  an  Abys- 
sinian, 99  ;  hiw  of  blood  for  blood 
among  the  Abyssinians,  119  ;  zoolo- 
gical division  of,  216,  221. 

Acanthodactylus  boskianus,  456. 

Acanthodactylus  mucronatus,  453. 

Accipiter  unduliventer,  55. 

Actitis  hypoleucos,  438. 

Actitis  ochropus,  443. 

Aden,  4 ;  sail  to,  ib.  ;  fauna  foimd  at, 
4,  6 ;  birds  shot  between  Bombay  and 
Aden,  5 ;  land  shells  found  at,  ib, 
rocks  of  volcanic  origin,  5,  193  ;  birds 
seen  at,  6  ;  author  leaves  Aden,  6 
return  voyage  to,  137. 

Adigrat,  march  to,  58  ;  geology  of,  59 
Adigrat  sandstones,  170. 

.£dicnemus  affinis,  428. 

iEdicnemus  crepitans,  428. 

iEdon  galactodes,  380. 

ifigialitis  niveifrons,  429. 

ifigialitis  tricollaris,  108,  429. 

^na  capensis,  65,  416. 

Agama  anncctans,  446. 

Agama  cyanogaster,  446. 

Agama  occipiralis,  449. 

Agula,  camping-ground  at,  64  ;  geology 
of,  i6. ;  niins  of  a  church  near,  65  ; 
scenery  of,  ih,  ;  departure  from,  66  ; 
oolitic  fossils  found  near,  99. 

Aiba  valley,  74,  76. 

Ajus4  coral  island  of,  reached,  6  ;  diffi- 
culty of  landing  on,  7  ;  description  of 
the  island,  ib.  ;  departure  from,  to 
Annesley  Bay,  ib. 

Alauda  arenicola,  387. 

Alauda  cristata,  387. 

Alauda  pnetennissa,  388. 

Alaudse,  46. 


Alcedo  semitorquata,  325. 

Alluvial  deposits  near  the  coast,  194. 

Amadina  polyzona,  408. 

Ammomanes  deserti,  390. 

Amydrus  albirostris,  46,  401. 

Amydrus  Blythi,  34,  899,  400. 

Amydrus  Hii])pelli,  46,  898. 

Anatidse,  family,  487. 

Ancylus  fluviatilis,  472. 

Animals    of   Abyssinia,  principles    of 
classifying,  214-221. 

Annesley  Bay,  7,  8;  country  around 
the  bay  and  its  fauna,  9-14. 

Anseba  valley,  fauna  of,  125. 

Antalo  limestones,  176;  description 
of  fossils  in,  201-203. 

Anthus  campestris,  46,  388. 

Anthus  cervmus,  46. 

Anthus  sordidus,  46. 

Aquila  rapax,  44,  295. 

Ardea  comata,  435. 

Anlea  gularis,  435. 

Ardeidse,  434. 

Ascida?,  family,  471. 

Ashangi,  lake  of,  77;  geology  of  the 
neignbourhood,  ib. ;  rear-guard  of  the 
army  at,  95 ;  x>hysical  geography  of, 
159. 

Atala,  valley  of,  73;  Abyssinian  in- 
dignation meeting  at,  73 ;  scenery 
of;  74. 

Auriculidffi,  family,  469. 

Aviculidse,  family,  471. 

Barbatxjla  pusilla,  34,  311. 
Bathyer^s  splendeus,  95,  279. 
Batrachia,  order,  458. 
Boni  Israel,  29,  81,  268. 
Bemicla  cyanoptera,  75,  85,  439. 
Birds, scarcity  of,  in  the  Indian  Ocean,  4. 
Bombay    Harbour,   author  leaves,  for 
Abyssinia,  3. 

I  I 


I 


483  ™ 

Roodon  lemniacatnin,  457. 

Brwlj-omU  cborolntina,  45,  95,  348. 

Bubo  cineraacciiH,  302. 

Rubo  lactcus,  802. 

BucciitideD,  403. 

Buemtaur,     tranxpnrt     ship,     anthor 

leaves  Bamljay  hnrboar  in,  for  Abjs- 

dnin,  3. 
Bucephalvui  typus,  458. 
Bucorrus  abysaini^us,  46,  830, 
Budftoa  i^Binpeatris,  382. 
BnilftcB  cinerooiapillB,  331. 
BuJ>-tea  iDcluiocepliBla,  13,  331. 
Bufo  panthvriDos,  459. 
Bulimua  labioaua,  5. 
Bulimua  Olivieri,  B5,  47S. 
Bullidm,  ratnily,  463. 
Buphags  erytlirorliyQcha,  46,  lOA,  401, 

402. 
Buteo  aagnr,  44,  K,  297. 
Bnyn,  camp  at,  49. 
Bjthinia,  Epodei  nova,  47X 

C. 

Calandbkll*  Anderasoni,  389. 
Calandrella  bmcliydaetyla,  12,  48,  389. 

Cunaroptara  brevicaudata,  378. 
Cania  meaomekg,  237. 
Cauis  Tariegatus,  44,  SS8. 
CapitonldfB,  family,  309. 
Caprimnlgidn,  family,  33S. 
Caprimulgna  inornatiiB,  337. 
Caprimttl^UH  nubicn^  336. 
Caprimolgtia  Uistigioa,  337. 
Cardiadn,  470. 
CamiTora,  ordsr,  228. 
Centiopiia  monacbns,  6S,  314. 
Ceatropua  supercilioms,  315. 
CepWophua  madoqtia,  68,  B3,  267. 
CercomMB  melonara,  13. 
CercointliecuB  griseo-viridis,  224. 
Cercotrichas  eiythropterus,  13,  360. 
Cerithiiilie,  466. 
CeHthiopeids,  468. 
Ceithilanila  desertorum,  12,  385. 
Cerjie  rudis,  326. 
Chamolio  IsTigatua,  128,  44I>. 
Chamidn,  470. 
Choiroptera,  order,  228. 
Cholidon  urbica,  249. 
Chelitropia  foimd  at  Adrn,  4. 
Chenalopex  lEgyptica,  46,  BJ,  438. 
Chcttusia  tnelanoptcra,  46. 
Chitonidffi,  46B. 


CUiia 


1,  316. 


Chmcliea,  AbTiniuan,  t9. 

Ciconia  abdintii,  436. 

CicooidK,  ^mily,  438. 

Ciniijs  belliaoa,  1S8,  4«. 

C^iu  KragiDosoa,  301. 

Circtu  dnencaos,  68,  SOI. 

CircOB  Swaiownii,  68,  SOI. 

CisticoU  hab«annica,  376. 

Clausilia  Sennaareiiau,  177. 

Climate  of  Al^HiiiM,  138. 

Coliidie,  family,  317. 

CDlitu  leucotiB,  its. 

ColiuH  macroums,  SI8L 

Culiuspasaer  flavi-acapniatna,  407- 

ColiuBpuser  UticaiidB,  405. 

Calamba  albitorquea,  46,  41S. 

Columba  guinea,  4ft 

ConidK  46S. 

Coraciaa  abyaainica,  96,  IIS,  319, 

Coraciaa  pUora,  SIV. 

Coral    formation    of    Bed    S«*  c 

196. 
CorbaUd*,  489. 
CorrLds^  family,  S93. 
Cormltur,  63,  804. 
Corrultur  cnaairoatria,  86. 
Corroa  affinia,  28,  127. 
Comis  Bcapnlatna^  13,  80,  SUS. 
Corythoniu  cjtuoKtiffiitt,  831. 
Coaaypha  Hmirola,  860. 
Cotyle  cincta,  96,  S40. 
Cotyla  minor,  860. 
Cotyle  rapeBtrii,  350. 
CrateropidK,  370. 

Orateropua  Imeooaphalw,  113,  $7'. 
Crateropoa  lencopyipn^  28,  1 4fi,  ] 
Crithagra  HaviTBrtel,  411. 
Crithngra  mfrnceps,  86. 
Critbi«™  Btnolata,  46,  41S. 
Cnculide,  family,  812. 
Oiicutus  canoruB,  812. 
Cynocephalns  hamai]lTa%  ISSL 
CypreiOE,  466. 
Cypselidie,  family,  334. 
Cypaelug  affinia,  336.  80S. 
Cypasliia  apoi^  8S6. 
CypaeloB  sqoatMialia,  4fi,  8W. 


Dartpeltis  abytainicna,  457. 
Denurlation  (in  Geology),  164,  16( 
Dir.runis  divaiii^atna,  28,  109,  341 
Dildi,  march  to,  81  ;  cfaanctor  ( 

TallcjB,  i6.  ;  camp  at,  ib. 
Dipta  gerbillua,  284. 
DJHBi,  island  of,  at  entraace  to  Ann 

Bay,  composed  of  tDetamorpbici 


INDEX. 


483 


Dolo,   caniping-grotind'  at,    66;   tele- 
graph put  up  at,  67. 
Dongolo,  valley  of,  63. 
Dorcas  gazelle,  14. 
Dromas  ardeola,  13,  142. 
Drymsecft  gracilis,  378. 
Drymceca  mystacea,  873. 
Drymieca  pulchella,  374. 
Drynuecie,  12,  45. 

E. 

EcHis  arenicola,  15,  457. 
Elephants,  wild,  115,  116. 
Emberiza  flayiventriu,  411. 
Eremomela  griseoflava,  355. 
£.strelda  minima,-  46,  409. 
Estrclda  pha^nicotis,  409. 
Estrelda  quartinia,  409. 
Estrelda  rhodopyga,  13,  408. 
Eudromias  asiaticus,  429. 
Eulimidse,  465. 

Euplectes  xanthomelas,  82,  93,  95,  405. 
Eupressis  Peronellii,  456. 
Eupressis  quinquetseniatus,  456. 
Eurystomus  afer,  320. 


F. 


Falco  barbarus,  288. 

Falco  sacer,  289. 

Falco  tanyptems,  93,  289. 

Falco  tinnuncnlufl,  290. 

Felis  caligata,  228. 

Felis  leo,  230. 

Felis  maniculata,  14,  226. 

Ferra  pass,  75  ;  geology  ofi  76. 

Finches,  128. 

Fissurellidfe,  468. 

Fokada,  author's  journey  to,  56  ;  geo- 
logy and  fauna  of,  57,  58  ;  camp  at, 
58. 

Formations,  geological,  list  of,  162; 
recent  fonnations,  194. 

Fossils,  description  of  the  new  species 
from  the  Antalo  limestone,  199- 
203. 

Francolinus  Erkelii,  28,  31,  46,  82, 
423. 

Francolinus  gutturalis,  46,  425. 

Francolinus  Kiippelli,  46,  425. 

FregiluB  graculus,  395. 

FringUla  tristriata,  413. 

Fringilla  zanthopygia,  413. 

Fringillidae,  family,  403. 

Fulica  cristata,  94. 

Fuligula  cristata,  94,  434,  437. 


O. 


Galerita  cristata,  12. 

Gallinula  chloropus,  94,  484. 

Gasteropoda,  462. 

Gastrochicnidce,  469. 

Gate  of  Weeping,  the,  6. 

Gazolla  lionnetti,  15. 

Gazella  Soemmeringii,  11,  16,  112,  260. 

Gazelle,  the  Dorcas,  14,  261. 

Gemitores,  order,  415. 

Genetta  tigrina,  233. 

€r«ological  formations,  list  of,  62. 

Geology  of  Abyssinia,  143-203 ;  la- 
bours of  previous  observers,  143-150 ; 
physical  geography  of  Abyssinia,  and 
its  relations  to  geology,  151. 

George,  Saint,  patron  saint  of  Abys- 
sinia, 49. 

Glarcola  pratjiicola,  431. 

Gongylus  ocWai^us,  456. 

Graculus  afiipenus,  94,  441. 

Graminicola  erythrogenys,  375. 

Graminicola  Le  Vaillantii,  375. 

Graminicola  robosta,  375. 

Grandala  leucogaster,  367. 

Greek  ruins,  49  ;  Greek  temple  near 
Agula,  06. 

Guinea-fowl  Plain,  29. 

Guna  Guna,  valley  of^  55 ;  scttjiery  of, 
66  ;  geology  of,  i6.  '   \ 

Gyogeranus  serpcntarius,  297. 

Gypaetps  barbatus,  44,  298. 

Gypastos  meridionalis,  25,  44. 

Gyps  Riigpelli,  44,  285. 


f 


••Nl 


H. 


Hadoda,  author  rides  over  to,  from 
Zulla  camp,  15  ;  trees  and  fauna 
met  with  on  the  jpumey,  tb,  ;  the 
camp  at,  described,  1(^  t  baboons  at, 
17 ;  geology  of,  17, 18  JJtAre  squirrel, 
Pectinator  Spckei,  18.  '^  *^; 

Halai,  49  ;  geology  of,  50-53.  " 

Halcyonida*,  322. 

Halcyon  semica^rulea,  322.         •>▼-••.. 

Halcyon  senegalemtis,  328.  *" "  ^^ 

Uarat  hills,  58.  \ 

Hari)iprion  carunculata,  85. 

Helix  Damaudi,  ^.,  474. 

Helix  pisana,  69. 

Helotarsus  ccaudatus,  44,  296. 

Herodius  garzetta,  435. 

Herpestes  mutgigella,  43,  234. 

Heteropoda,  class  of  Mollusca,  462. 

Hinmuinidae,  family,  346. 

Hirundo  lethiopica,  347. 


X 


■  -v. 


I 


\ 


484 

Hinmdo  ilpwitrU,  846. 

Hirondo  melanocriBiaa,  SW. 
Hirundo  puella,  S18. 
Hirundo  niEcBp*.  348. 
Himndo  nutic&,  317. 
Homalotonu  latrix,  ifiS. 


Hjnna  crocota,  14,  296. 
Hfziu  atrUU,  14. 
Hyaeiua,  68. 

HypbaotoniU  nlbols,  IS,  tOt. 
Hypb&ntonui  Guerini,  40. 
HyplialitoroiB  Itrnta,  403. 
HTplLantaniil  loteola,  404. 
HTpluDtomU  mtluiotu,  403. 
HTpoIau  elucs,  S80. 
Hypolaii  languida,  379. 
HypotiioTchu  ooncolor,  93. 
Hyisx  abywiniciu,  249. 
Hyntx  Bnicei,  43,  262. 
Hf»z  t  apeciea  nova,  857. 

I. 
Iahthiha,  three  tpecies  of,  fuQDi]  *t 

Aden,  4. 
Ibia  caiuncnlsta,  9S,  437. 
Ibia  comata,  97,  43(1. 
IbiH  HupiprioQ,  46. 
Tndinu  Ocean,  ecarcit?  of  birds  in,  4. 
Indicatoridee,  tamity,  807. 
iDdicstor  minor,  S07. 
Indicator  Spamunai,  307. 
InMHBores,  order,  305. 
Jriaor  aterrimiu,  334. 
IriaoT  erythrothynchua,  45, 109,  332. 
lapidina  picta,  323. 


Jackals,  foand  at  Znlla  or  Halkatto, 
14  ;  probably  tbe  Canit  ritMriat  of 
Hemprich  and  Ehrenberg,  ib. 

Jitta  rarins,  87  ;  geology  of,  S3. 


Kblliada,  470. 

Eereo,  Tiiiit  to,  123. 

Kokai,  115  ;  wild  elephants  at,  ib. 

Kolqual,  rare  plan^  fomid  at  Mayen, 
29. 

Eonmyli,  anthor's  ride  over  to,  fh)in 
Hadoda,  21  ;  wslla  dug  tliere,  21, 
22 ;  geology  of,  23 ;  roods  made 
there,  24  ;  people  of,  25 ;  their  clia- 
ncter,  ib.  ;  and  food,  it.  ;  fauna  of, 
26, 38  !  two  days'  stay  at,  38  :  heary 
&U  of  rain  at,  i6. 


Lackbta  mnharica,  449. 
Lacerta  Storti,  463. 
Lamellibnncluata,  4  SB. 
I^mprocoliiu  chalybtena,  67,   84, 

109,  396,  396. 
lAmprocolina  chiyiogaatar,  128,  3! 
Lamprotomia,  46. 

Lamprotornia  pnrpnToptera,  1S6>,  3 
Land  sheila,  38. 
Laniada,  family,  837. 
Lauiarios  craentotas,  13. 
lAniarina  erythropteriu^  45,  343. 
Lanina  coUorio,  340. 
Idknina  crnentns,  342. 
I«niuB  &llai,  337. 
lAniuB  gambenaia,  842. 
Lanins  hnmeralis,  45,  338. 
LaniuB  isshelUnua,  12,  S3ff. 
Iiauios  lahtora,  1 2. 
Lanins  rufoi,  340. 
Uridn^  440. 
Lania  macns,  13,  440. 
Irna  Hempriclui,  IS,  441. 
Iatub  loncophthalmns,  18,  441. 
Lebka  Talley,  the  march  tluxingb, 
Lepns  tegypticns,  10,  276. 
Lepos  ti)^Bsis,  128,  274. 
Limestones  of  Antalo,  176,  ISO. 
LioDSRB  shot,  133. 
Littorini4«,  467. 
LobtTanellas  melanocephaliu,  86, 

430. 
LohinineUns  »enq(alas,  431. 
Lacinids,  470. 
Lynmea  Natatenna,  472. 


Maogiuivbata  fonnd  at  Aden,  4. 

Macronrx  BsTicolIis,  85,  93,  384. 

Uactrids,  409. 

Magdaia,  arrival  before,  89  ;  proi 
cistion  of  the  word,  A.  ;  geolof^ 
the  neighbonrbood,  90  ;  camp  at, 
author  departs  from,  92. 

Mai  Wahiz,  QO ;  camping-gronad  at, 
geology  of,  ao,  63. 

MalttcocercuB,  373. 

Halkstto,  see  Zolla. 

Manimatia,  221-284. 

MargenolliJip,  family  of  MoUnaca,  4 

Maangita,  campiTic-groiind  at,  78. 

Usyon  or  Uniful  Wells,  28  ;  fauna 
28,  29,  34  ;  exploration  of  the  Ul 
Valley,  30  ;  search  ftfr  springi^  3 

Melania  tubercolata,  472. 


INDEX. 


48-5 


Melierax  polyzonus,  2dl. 

Meliphagidse,  family,  354. 

Merops  albicollis,  821. 

Merops  eiythropterua,  321. 

Merops  Lafresnajd,  28,  322. 

Merops  nubicus,  13,  321. 

Merops  superciliosus,  321. 

Merops  yiridissimus,  13,  320. 

Meshek,  scenery  and  geology  of^  71. 

Metamorphics,  164-169. 

MilvidsB,  464. 

Milvns  ffijprptius,  44,  300. 

Milvus  migrans,  13,  44,  300. 

Mollosca,  462. 

MoUusca,  oceanic,  462 ;  marine,  ob- 
tained from  Annesley  Bay,  463 ; 
freshwater,  472  ;  land,  473. 

Motacilla  alba,  13,  380. 

Motacilla  solphurea,  381. 

Mus  abyssinicas,  82,  85,  93,  283. 

Mas  dembeensis,  284. 

Mns?  species,  284. 

Mytilidse,  470. 


N. 


Natatores,  437. 
Katicidse,  family  of  Mollusca,  464. 
Nectarinia,  13. 
Kectarinia  afSnis,  45,  351. 
Nectarinia  cnientata,  45. 
Nectarinia  liabessinica,  351. 
Nectarinia  Jardinii,  352. 
Nectarinia  metallica,  353. 
Nectarinia  pulchella,  354. 
Nectarinia  Tacazze,  45,  82,  352. 
Nectarinidse,  family,  351. 
Neophron  percnopterus,  13,  287. 
Neophron  pileatus,  18,  287. 
Neotragus  saltianus,  11,  125,  268. 
Neritidae,  467. 
Nilaus  brubru,  108,  344. 
Nisus  badins,  114. 
Nisus  niger,  119,  293. 
Nisus  niloticns,  108,  292. 
Nisus  sphenums,  114,  294. 
Nisus  tachiro,  291. 
Numenius  arquatus,  432. 
Numenius  phseopus,  432. 
Numida  ptilorhyncha,  10,  46,  421. 
Numidids,  421. 

0. 

Obsidian,  flakes  of,  197. 

Olidae,  427. 

Oligocercus  micraras,  85,  376. 


OlividflB,  464. 

Oolitic  fossils  found  near  Agula,  99. 

Ophidia,  order,  457. 

Opisthobranchiata,  468. 

Oryx    beisa,    113 ;    one    shot,    135  ;■ 

scarcity  of,  135,  262. 
Oryx  leucoryx,  136. 
Orootragus  saltatrix,  265. 
Oriolidse,  369. 
Oriolus  monacha,  369i 
Ostreidse,  471. 
Otis  Arabs,  10,  109,  427. 
Otis  aurita,  69. 
Otis  melanogaster,  57,  427. 
Otis  Rhaad,  46. 
Oxylophus  afer,  81 2. 


P. 

Pachyderm  AT  A,  order,  241. 
Palseomis  torquatus,  305. 
Paridap,  family,  856. 
Parus  leuconatus,  356. 
Parus  leucopterus,  356. 
Passer  canicapillus,  412. 
Passer  pyrjjita,  413. 
Passer  Swainsoni,  46,  109,  128,  411. 
Passeres,  sub-order,  837. 
PatelUdae,  468. 
Ppctinator  Spekei,  18,  281. 
Pelicanidse,  family,  442. 
Pelicauus  phillipensis,  442. 
Pelouiedusa  gehafie,  444. 
Perdicidaj,  423. 
Perifitera  afra,  108,  417. 
Petrocincla  cyanea,  357. 
Petrocincla  rufo-cinerea,  358. 
Petrocincla  saxatilis,  358. 
Phacochffirus  iEliani,  or  wart-hoff.  10, 

241,  249.  ^ 

Phaetonidfle,  family,  441. 
Phylloscopus  abyssinicus,  878. 
Phylloscopus  umbroyirens,  878. 
Physa  contorta,  472. 
Pica,  sub-order,  805. 
Picid®,  family,  305. 
Picus  Hempnchii,  34,  306. 
Picus  nubicus,  34,  305. 
Pionns  Meyeri,  114,  804. 
Pisces,  class,  460. 
Planaxidffi,  467. 
Planorbis  Ruppelli,  472. 
Platysteira  pririt,  45,  345. 
Podiceps  auritus,  440. 
Podiceps  cristatus,  94,  440. 
Podiceps  minor,  440. 
Pognornynchjis  abyssinicus,  309. 


486 


INDBJL 


Pognorhynchns  melanocepbalos,  810. 

Pognbrhynchos  undattis,  46,  810. 

Pratincola  Hemprichii,  364. 

Pratincola  pastor,  864. 

Pratincola  nibicola,  864. 

Pratincola  semitorqoata,  95,  866. 

Pratincola  sordida,  82,  98,  866. 

Psalidoprocne  pristoptera,  78,  840. 

Psamnophis  sibilans,  128. 

Psittacula  Tarantffi,  804. 

Ptemistes  robricollis,  10,  16,  109,  426. 

Pterocles  exnstus,  419. 

Pterocles  gutturalis,  64. 

Pterocles  Lichtensteinii,  11,  419,  421. 

PteroclidsB,  419. 

Ptcropida,  462. 

Puhnonifera,  469. 

Pupa  insularis,  6. 

Pnpee  Senegalensia,  ftc.,  476. 

Porpuridffi,  464. 

Pycnonotidse,  869. 

Pycnonotus  arsinue,  16,  46,  869. 

I^rrhulauda  albifrons,  891. 

Pyrrhnlauda  melanauchen,  890. 

I^teUa  citerior,  18,  410. 


Q.    - 

QrADBUMANA,  order,  222. 
Quelea  sethiopica,  108,  406. 
Qnerquedula  crecca,  488. 

R. 

Rahaotjddt,  camping-ground  at,  86. 

Ballus  Hougeti,  68,  94,  488. 

Rana  madagascariensis,  468.    * 

Raptorial  birds,  126. 

Rasores,  419. 

Religions  of  Abyssinia,  118,  119. 

Reptilia,  class,  446. 

Rhinoceros*  bunt,  121. 

Rbizomys,  96. 

Rbyncbeea  bengalensis,  482. 

Rissoidse,  467. 

Rocks,  volcanic,  at  Aden,  190-198. 

Rodentia,  order,  273. 

Romanos,  Saint,  his  shrine,  66. 

Ruminantia,  260 ;  domestic,  272. 


S. 


Salt  used  as  small  change,  67. 
Sandstones  of  Adigrat,  170-176. 
Sarciophorus  tectus,  11,  128,  430. 


Sanria,  order,  445. 
Saxicola  Ananthe,  46^  861. 
Saxicola  frenata,  85,  08,  861. 
Saxicola  isabellina,  45,  361. 
Saxicola  lugena,  46,  868. 
Saxicola  lugubris,  46,  868. 
Saxicola  melanora,  868. 
Scolopacido^  fiunily,  480. 
ScopophoruB  montanns,  9^ 
Scops  senegalensia,  808. 
Senafi6,  36 ;  fauna  of,  85,  48  ;  geolo 

of,  86,^89,  66;    weatliAr    Sk,   8 

people  of^  47. 
S^inus  cintrinelloidea,  414. 
Serinufl  nigrioeps,  415. 
Shells,  land,  88. 
Shohos,    robbery  oommittad  by,  9 

Shoho  Tillage,  82. 
Sinusigera,  4. 
Snakes,  128. 

Soils  of  the  Abyssiniaii  ^*g^^n*^^^  Ij 
Solaridse,  466. 
Spondylidai,  471. 
Stellio  cyanogaster,  66. 
Stenoj^yra  gracilis,  476. 
Strepsiceros  Kudu,  270. 
StrombidsB,  466. 
Stumidse,  896. 
Succinea,  476,  476. 
Suru  pass,  26  ;  flood  at,  27 ;  camp 

Suru,  ib, ;  fauna  of^  28. 
Sylvia  dnerea,  879. 
Sylvia  melanocephala,  870. 
Sylviadae,  876. 


T. 


TakonDA,  68. 

Tantalidee,  486. 

Taranta  pass,  47. 

Tchitrea  melanogastra,  844. 

Telegraph  set  up  at  Dolo,  67. 

TellinidflB,  469. 

Terebridae,  466. 

Terekia  cinerea,  438. 

Thamnolsa  albiscapulata,  860. 

Tockus  erythrorhynchus,  16,  108,  321 

Tockus  flavirostris,  28,  46,  826,  820. 

Tockus  Hemprichii,  46,  827. 

Tockus  nasutus,  113,  829. 

Trachyphonusjnargaritatua,  16,  311. 

Trappean  series,  181. 

Tringse,  433. 

Triton idse,  family  of  MoUusca,  463. 

Trochidae,  467 

Turacus  leucotis,  316. 

Turbiuidaj,  468. 


f 


INDEX. 


487 


Turdus  oliyacinnfl,  45,  95,  357. 
Tardus  simensis,  45,  357. 
Turtur  lugens,  46,  416. 
Turtor  semitorquatus,  416. 
Turtur  senegalensis,  417. 

U. 
dNDiTL  Wells,  au  Mayon. 

V. 

Yaraihts  ocellatus,  445. 
Veneridffi,  470. 
Yespertilio,  226. 
Vidua  Serena,  128,  408. 
Vidua  sphenura,  407. 
Volcanic  rocks  at  Aden,  5, 190. 


W. 


Wadela  plateau,  fauna  of,  85. 
Wandig  pass,  83. 
Wells,  21-22,  28,  28,  87. 

X. 

Xesub  leuco-umbrinus,  44,  279. 
Xerus  rutilus,  26. 

Z. 

ZONITES,  475. 

Zoology  of  Abyssinia,  its  bibliography, 
208-218. 

Zosterops  abyssinica,  855. 

Zosterops  poligastra,  854. 

ZuUa  or  Malkatto,  camp  at,  8,  87  ;  vol- 
canic character  of  hills  at,  i&. ;  excur- 
sion from,  9  ;  fauna  of,  12-15. 


THE  END. 


II 


I 


I 


TATUJB,    PUKTXU, 


IJf 


x^C 


H